Corporate social responsibility and the supposed moral agency of corporations


Corporate Social Responsibility (CSR) has been traditionally framed within business ethics as a discourse attempting to identify certain moral responsibilities of corporations (as well as get these corporations to fulfill their responsibilities). This theory has often been normatively grounded in the idea that a corporation is (or ought to be treated as) a moral agent. I argue that it is a mistake to think of (or treat) corporations as moral agents, and that CSR’s impotency is a direct result of this mistake. I then outline a distinction between business ethics and business politics, arguing that CSR might be better framed as a political goal – one which might be able to take better advantage of the resources of corporate governance and a renewed (albeit shifted) focus on agency theory.


The classic – and still largely dominant[1] – account of corporate governance has it that governance ought to operate in the interests of the stockholders, and that, in Milton Friedman’s (1970) words, ‘the social responsibility of business is to increase its profits’. This ought is grounded primarily legally: the executive has a contractual fiduciary responsibility to the shareholders as their ‘agent’. David Ciepley, in an article thoroughly critical of agency theory, still rightly recognizes that this is not simply a theoretical issue, but also a matter of de facto law: Ciepley (2013: 154) traces a history of US legal precedents across ‘the long nineteenth century, and with renewed energy since the 1960s’, treating corporations as the private property of their stockholders, and treating executives as agents with fiduciary responsibilities to pursue the desires of those stockholders.

Critics of this dominant approach who want to offer competing theories of corporate governance – especially in light of the legal entrenchment of agency theory – typically do so on the basis of pragmatic and strategic claims. Blair and Stout (1999), for example, argue that the dominant principal-agent theory is wrong on two major counts. First, they argue that corporate management – and even corporate law – doesn’t really work the way agency theory says it does. And second, Blair and Stout claim that management ought to operate for the benefit of the corporate entity itself, rather than for the benefit of shareholder wealth maximization, because it is in the financial long-term interests of the corporation. In other words, agency theory is both factually wrong and less strategically beneficial. Likewise, Rajan and Zingales (2000) also make the claim that the corporation ought to be governed in the interest of the corporate entity itself (‘the firm’), rather than in the interests of the shareholders. They ground this claim by an appeal to ‘the facts’: the nature of modern corporations has changed, claim Rajan and Zingales (their chief example is the modern investment firm, a very different kind of enterprise than mid-20th century General Motors was), and to govern in the old way will simply result in failure (bankruptcy, dissolution of the firm, etc.).

As competing theories of corporate governance, theories of Corporate Social Responsibility (CSR) can also object to the dominant approach on both pragmatic, factual and strategic grounds. The latter (the ‘morality pays’ approach) offers a competing goal of corporate governance (a corporation ought to be governed, say, in the interests of all of its stakeholders), and justifies this claim on grounds similar to those used by Blair and Stout or Rajan and Zingales: either that this kind of governance will make the firm more successful, or that failure to govern in this way will cause the firm to fail. We shall return to this approach in a moment. The other major strategy in CSR arguments, meanwhile, will be to claim that (as a matter of fact) the corporation carries with it some sort of social obligations, usually in the form of moral obligations. It is not uncommon to see both strategies employed side by side.[2]

There are two major problems with the ‘morality pays’ justification for CSR. First, as John Corvino (2006: 4ff.) has argued, the ‘morality pays’ version of business ethics fails in precisely those situations in which it is most needed. (In other words, it is just those situations in which doing the right thing does not pay – or, worse, that doing the wrong thing is better for the business – that business ethics is most needed.) Of course, one might simply respond that such ‘hard cases’ don’t exist; doing what is right and doing what is profitable may be separate in theory, one might say, but at least they are the same thing in practice. This, however, is the second major problem with the ‘morality pays’ argument: it simply isn’t true. Henry Mintzberg (2007: 38-41), for example, cites some rather compelling empirical evidence which suggests that the most profitable corporations are the ones which place the least emphasis on ‘social responsibility’ – and (more damningly) vice versa.

Without strategic justification for CSR policies, theorists must ultimately resort to arguing that corporations (or executives) have social obligations, even if these obligations sometimes stand in the way of profits. There seem to be two major ways to make this claim. One can, starting with the moral responsibilities of human beings in general, attempt to show how these duties make certain moral claims on the practice of business executives (ones which trump their other role responsibilities). Or, one can attempt to ground CSR and calls for corporate ‘citizenship’ on an idea that the ‘business organization’ itself ought to be thought of as an acting, ethical subject. The idea here seems to be that if a corporation is a moral agent, then we can make certain non-(or extra-)economic demands on it to fulfill certain social (i.e. moral) responsibilities.

Because the field of CSR is wide, my goal in this article will not be to pay special attention to any particular theorist’s work – rather, I shall try and focus on the general principles that must be at stake in putting forward CSR as a goal of corporate governance. In the pages that follow I will argue that grounding CSR in ethics is ultimately a mistake (or, rather, that it invariably masks a confusion). After I have laid out my objections, I will return to corporate governance and agency theory to think about some of the resources still left to those of us who are interested in CSR. I will argue that, while CSR does not ultimately make sense as an ethical claim, it can (and should) be thought of as a political goal. Furthermore, I will outline a form of Corporate Social Responsibility that does not compete with agency theory, but instead harnesses it. My ultimate goal will be to suggest a shift away from ‘business ethics’, toward ‘business politics’. In the final section of this essay, I will both explain what I mean by this and attempt to address some preliminary objections to my position.


Theorists who want to underpin CSR with ethics attempt to show that corporations have certain moral responsibilities to strive for some social ends over and above the pursuit of profit. However, the attempt to show that such moral responsibilities exist can be done in either (or both) of two major ways. The first approach – seen, for example, in Kenneth Goodpaster’s ‘principle of moral projection’ – is to treat a formal organization as an entity capable of deliberation, making decisions, and holding moral values. The second approach is to try and establish ethical constraints that apply specifically to executives, managers, and/or other official representatives of a company qua their roles as official representatives. (This might be a matter of showing that executives have special, extra duties – duties that they have over and above their duties as human beings – or it may simply be to show that moral duties in some way ‘trump’ any occupational duties.) These two approaches to establishing moral obligations for corporations are not mutually exclusive – Goodpaster (2007), for example, attempts to pursue both lines of argument[3] – but nor does one necessarily entail the other. Therefore, we will (briefly) treat each one separately; in sections two through five, I will consider the various ways in which one might try to treat businesses (primarily corporations) as moral entities. After a brief detour (in section five) in which I consider the roots of this mistake, I will turn ­– in section six – to the idea that business executives have special moral responsibilities, and set this idea back into its context within agency theory.

The attribution of agency to business organizations – especially corporations – is certainly commonplace in everyday speech. To say that ‘General Motors is reopening a plant in Texas’ or that ‘Lehman Brothers misled its investors’ is, grammatically speaking, unlikely to cause confusion. But neither, it should be noted, are such sentences as ‘My car is burning fuel’ or ‘Germany invaded Poland’. The simple fact that we can attribute agency without causing grammatical confusion, in other words, is not enough to show that we can attribute personhood to a subject, let alone moral personhood. A car can certainly take the subject-position in a sentence, but it cannot be the subject of even legal rights, let alone moral responsibilities.

The argument that a corporation has moral agency can, it seems to me, proceed along any (or all) of three major lines. First, one can argue that a corporation is capable of taking intentional action, and is therefore an agent subject to moral address and sanction. Second, one can argue that, even if corporations are not (and cannot be treated as) moral agents in the primary sense, that they might still be considered ‘secondary moral agents’. Finally, one can argue that, even if corporations do not meet the metaphysical standards of personhood, it might still be useful to treat them as moral persons. Each of these arguments attempts to establish a normative basis for Corporate Social Responsibility on the claim that a corporation is (or should be treated as) a moral person. As none of these arguments are logically reliant upon any of the others, allow me to address them individually, in turn: In section three I shall argue that a corporation is not a moral person; in section four I shall argue that the notion of corporations as ‘secondary moral agents’ is both wrong and unhelpful; and in section five I shall argue that it is not useful to treat corporations as if they were moral agents.


The claim that a corporation – or, indeed, any formal organization – is a moral agent always seems to build from the argument that corporations are intentional agents. As French (1979: 211) understands the issue, for a corporation to be a moral agent ‘it must be the case that some things that happen, some events, are describable in a way that makes certain sentences true, sentences that say that some of the things a corporation does were intended by the corporation itself’. The key distinction for French, then, is whether or not ‘attributing intentions to a corporation is only a shorthand way of attributing intentions to the biological persons who comprise, e.g., its board of directors’ (ibid.). To this end, French devotes considerable effort into developing the concept of a ‘CID Structure’ (Corporation’s Internal Decision-making Structure) in order to show that decisions made by a corporation often cannot be reduced to (or redescribed as) decisions made by specific biological persons employed by that corporation.

There are some very important debates about how corporate intentionality actually works.[4] Note, however, that such debates miss a more important point; for even if we fully accept some account of group constitution such that it would count as a theory of how ‘corporate decisions’ get made, and even if we accept fully that this account establishes the point that corporations are intentional agents, a more fundamental problem still stands. The jump from arguing that a corporation is an intentional agent to the claim that a corporation is a moral agent rests on the premise that all intentional agents are moral agents. And this claim is certainly false.

First and foremost, it should be apparent that intentional agency is far too broad a criterion to identify only those subjects that are moral persons; John Danley (1980: 144), for example, draws our attention to the case of chess-playing computers. While it is certainly intelligible to say that ‘the computer intends to respond P-K4 to my king pawn opening’, it would by no means be correct to conclude from this that the computer is a moral person. More clearly still, Daniel Dennett uses the example of an apple tree. ‘You can trick an apple tree into “thinking it’s spring”’, he writes, ‘by building a small fire under its branches in the late fall: it will blossom’ (1988: 149). It is important to note that this is not simply a metaphorical extension of the notion of intentionality, nor a reductio ad absurdam; ‘intentional agency’ is a stance whereby we attribute certain beliefs, desires, and intentions to an entity in a way that allows us to both ascribe certain actions (blooming; moving a chess piece) to that entity and explain those actions by reference to intentional states (thinking it’s spring; responding to my king pawn opening). Furthermore, as Dennett (ibid.) notes, we either treat intentions as something we ascribe to an entity (Dennett calls them ‘Intentional Systems’) – that is, a certain stance we adopt towards them – or we get bogged down in potentially intractable metaphysical debates about ‘real’ versus ‘false’ intentions. In this sense, intentionality is a useful concept inasmuch as it allows us to predict and describe behavior in the aforementioned ways.

There may be in every case other ways of predicting and explaining the behavior of an Intentional system – for instance, mechanistic or physical ways – but the Intentional stance may be the handiest or most effective or in any case a successful stance to adopt, which suffices for the object to be an intentional system. (Dennett: ibid.)

We might go on to note that, while ‘the Intentional stance’ seems to apply just as well to persons, corporations, and apple trees, the sort of metaphysical claims that might be made to make ‘intentionality’ only apply to moral agents (e.g., intentions as reflective states of consciousness) will, even if successful, rule out corporations just as much as they will rule out computers and plants. The mistake of those who equate intentional systemhood with moral personhood is understandable, as intentionality is a necessary condition for moral personhood. The mistake, however, is in assuming that intentional systemhood is also a sufficient condition for moral personhood.[5]

It might be thought that a better case might be made for corporate moral agency by appealing to rational agency instead of strictly to intentional systemhood. Following List and Pettit (2006), we could show that only certain constitutions (CID structures) will result in rational agency at the level of the group, and we could argue that only these groups are moral agents. Restricting ourselves to rational agents would seem to allow us to treat only some intentional agents as moral agents, while excluding cases like those mentioned by Dennett: dogs, ivy plants, and apple trees. However, it should again be clear that any definition of ‘rational agency’ sufficiently flexible to include corporations would also have to include chess-playing computers and the like. In other words, either our understanding of what it means to be a rational agent is sufficiently restricted that it applies only to moral agents (and, through this restriction, excludes corporations); or our understanding is broadened to include corporations, but in the process becomes too broad to apply only to ‘persons’. It is for this reason that Amelie Rorty (1973) has argued that those who equate rational agency with personhood ‘must include many other intellectual capacities – memory, imagination, perception, perhaps even the formation of desires – within the analysis of rationality’ (ibid.: 71). Rorty’s argument makes another crucial objection, though. She in fact argues that, even when we apply it only to bodily agents, the theory that persons are rational agents ends up being too narrow to properly define personhood (ibid.: 79). In other words, even if we were to broaden our understanding of rational agency such that it might include corporations (and chess-playing computers), it is likely that we would yet fail to include less contentious candidates for moral personhood! For example, what human being can be said, in practice, to ever be fully rational? We might instead treat ‘rationality’ as a normative ideal – ‘a person is identified as someone for whom the ideal of rationality can be dominant’ – but only at the cost of making rationality, as Rorty shows, ‘at best a necessary but not a sufficient condition for being a person’ (ibid.: 69).

But if intentional systemhood and rational agency are not sufficient conditions for personhood, then what is missing? Dennett (1988: 147-148) offers what I believe to be three very reasonable criteria. Not only must a person be an entity toward which we can take an intentional stance (that is, attribute intentions to), but a person must 1) be capable of reciprocating this stance (that is, must be a ‘second-order Intentional system’; ibid.: 151); 2) be capable of verbal communication; and 3) be self-conscious. Dennett himself goes on to make the point that these conditions for personhood (Intentional systemhood, plus the three additional requirements) are necessary, yet still not sufficient conditions of personhood. And why? Because these criteria are normative – and we are always (especially in the most important cases) unsure about what counts as a ‘passing grade’ (ibid.: 163). All the same, it should be clear that these criteria easily apply to our least contentious cases of moral persons – ‘normal’, adult human beings – while easily excluding goldfish, ficus plants, chess playing computers – and corporations.

Two objections still seem to be available to those trying to insist that a corporation can be a moral agent. First, one might ask: but why can’t moral agency supervene at the group level the same way that rational agency can? And second, one might object that I have been laying out the conditions of moral personhood, and ask if it isn’t possible for a corporation to be a moral agent without being a moral person. And so, before I move on to consider the idea that corporations might be secondary moral agents, let me first briefly address these objections to my dismissal of the idea that corporations are primary moral agents.

The account of the supervenience of rationality at the group level depends, essentially, on building rationality into the decision-making structure up front. List and Pettit (2006: 96) give an example they call the ‘premise-based procedure’ for making group decisions, in which logical constraints are built into the procedure itself (and accepted by all group members), asking group members simply to decide individually on premises, and then allowing logical conclusions to be derived from these premises. The group constitution contains a kind of logical computer: given inputs, it’s ‘programmed’ to calculate the logically-necessary results. It’s certainly not simple to make rationality supervene at the group level – this, after all, is List and Pettit’s main point – but rationality can supervene by simply building rationality into the decision-making rules; ‘rationality’, after all, is simply (in this sense, at least) a matter of logical self-consistency. It accepts any input at the level of premises, any input at the level of goals; a conclusion is rational simply to the extent that it efficiently and self-consistently identifies conclusions and means to the given ends.

It may seem that we could simply build morality into the decision-making structure in the same way – say, by disallowing in advance certain kinds of means, and even certain ends. We might build a requirement for morality into the decision-making structure by fiat – but would this be enough for moral agency to supervene at the group level? There are good reasons, I think, to answer in the negative. Again, the example of a computer program is helpful: we could program a robot to follow certain rules (moral rules, cultural norms, etc.), ‘building morality in’ in advance. But would we really say that this robot had moral agency? The robot would in no way be regulating its own conduct; it would not have any self-awareness, and it would not have free will. (In Kant’s terms: for such a robot, moral laws would be descriptive laws, merely telling us how the robot will act, rather than normative laws, telling us how the robot ought to act.) We would have good reasons to prefer spending time with morally perfect robots and corporations, I think – but these would still not be fellow moral agents. Morality simply cannot supervene at the group level in the same way that rationality does.

But the objection will still seem to be open that I am setting the bar too high, demanding that we only count moral persons as moral agents. A number of ‘improvements’ on French’s account have recently been offered (see Manning, 1984; Arnold, 2006; or Hess, 2013), arguing that even if a properly-constituted CID structure wouldn’t make a corporation a moral person, it still might make a corporation a moral agent. Manning (1984) and Hess (2013), for example, both seem to reason that intentional, rational agency is enough for moral agency, even if it is not enough for moral personhood. Hess (2013: 335) writes that ‘personhood… does not follow from moral agency’, per se, but seems therefore to think that moral agency is a necessary-but-not-sufficient requirement for moral personhood. But if there is any difference at all between rational agency and moral agency (and, looking again to our robots, we have good reason to think that there is), it would seem that arguments like Hess’s and Manning’s have things exactly the wrong way around: moral personhood does not follow from moral agency, because moral agency follows from moral personhood! In other words, the difference between a rational agent who can be morally responsible and a rational agent who cannot is going to rest on precisely those features (self-awareness; ability to take up second-order intentional stances; etc.) that mark out moral personhood. Moral agency – the ability not only to take intentional action, but also to be morally responsible for one’s own actions – is a special feature of moral persons, and not the other way around. If a corporation is not a moral person, then it is also not a moral agent.


But granting that corporations are not true (‘primary’) moral persons, those looking to ground business ethics in the moral agency of businesses might yet try to argue that corporations are ‘secondary moral agents’. This is the claim that has been put forward by Patricia Werhane (1985), drawing on a theory of secondary actions developed by David Copp (1979).[6] Allow me to lay out this argument briefly and clearly as I understand it.

A corporation, as we have said above, is an intentional system, but not a person. As such, it is capable of ‘acting’, but not directly; rather, the corporation’s ‘actions’ must be the actions of agents of that corporation on the corporation’s behalf. Through whatever process is required to ‘incorporate’ certain actions of employees of a corporation as the ‘corporation’s actions’ – and here Werhane (1985: 54-55) essentially appeals to a version of the ‘CID Structure’ or ‘constituion’ – corporations become at least indirectly responsible for these actions. In Copp’s (1979) language, the actions of the agents are ‘primary actions’, and these primary actions are – under certain conditions (the CID Structure, etc.) – said to constitute the ‘secondary actions’ of the corporation. Both Copp (1979: 177) and Werhane (1985: 52-53) compare the situation to a person using a real estate agent to rent or purchase property; the agent takes a primary action, but does so on behalf of the client, who in this way takes ‘secondary action’. I am therefore not wrong when I say, ‘I bought a house’, for the agent was merely acting as my representative. However, both Copp and Werhane also note a crucial distinction: whereas in the case of a person acting through an agent to buy property, both parties are persons (and, hence, moral agents), in the case of a corporation ‘acting’ through its representatives, only the representatives are self-sufficient agents. Because a corporation is not a self-sufficient agent, Copp (1979: 185) points out, it cannot be a self-sufficient moral agent. However, it is at this point that Copp lets the crucial issue slip past him:

That is, simply, the actions of a collective cannot accord, or fail to accord, with the requirements of morality unless some person performs an action which constitutes the collective’s doing so. (ibid.)

Because a person is a moral agent, his or her actions can accord – or fail to accord – with the requirements of morality. But if a corporation is not a moral agent, then how can there be requirements of morality for its actions to meet or fail to meet? Copp immediately goes on to say, ‘It should not be assumed, however, that “moral attributes” simply transfer across the constitution relation’ (ibid.). But while this is absolutely correct, he gets the inference exactly wrong: ‘It should not be assumed, for example, that if a collective is blameworthy for something it has done… then that individual is blameworthy for this given action’ (ibid.). In other words, after noting that a collective is not a moral agent, and that moral attributes do not automatically transfer across the constitution relation, Copp goes on to assume (simply and without argument) that the actions of a collective can still be morally obligated.

Though she does not criticize him on these grounds, I believe that Werhane notices the missing step in Copp’s argument. She therefore attempts to explain how moral praise and blame can transfer across the constitution relation. However, her explanation falls short of exactly being an argument: ‘because secondary actions are, in a derivative way, actions of persons’, she writes, ‘they can be moral or immoral actions, and one may evaluate them accordingly’ (1985: 57). Where Copp was right to note that morality does not simply transfer across the constitution relation, Werhane attempts to circumvent this clause by saying, in effect, ‘but morality transfers anyway’. It is hard to see this as anything but a fallacy of composition: the actions of corporations are composed of the actions of moral persons, so therefore the actions of corporations are moral (or immoral) actions. Rather than further argument, Werhane attempts to appeal to examples; but these examples should only serve to bring us back to the points we have been making all along. The example of Nestlé marketing formula in the third world is used to show that corporate misdeeds provoke moral outrage (ibid.), while the example of Johnson & Johnson retracting Tylenol during the 1982 scare is used to show that corporations (sometimes) respond to moral demands (ibid.: 58). But neither public outrage at corporate misdeeds nor the fact that corporations are capable of responding to consumer pressures are enough to make intelligible the claim that a corporation is any kind of moral agent, primary or otherwise. In the end, it is unclear how Copp and Werhane are doing anything other than simply redescribing intentional systems as moral agents.


A corporation, then, is not in any sense a moral person. One might still claim, however, that it is useful to treat corporations as if they were moral persons. Kenneth Goodpaster, for example, in relying upon the ‘principle of moral projection’, tends to treat this principle as an axiom rather than a thesis to be defended. In Conscience and corporate culture the principle is asserted as a sort of heuristic – Goodpaster cites the model of Socrates using the city as an image of the soul writ large in Plato’s Republic as one of his precedents (2007: 19) – and it would seem that, in the end, Goodpaster’s argument in that book depends upon whether or not his analogy proves helpful as a way of understanding business ethics. Inasmuch as it tends to contribute to a fundamental confusion of normative grounding, I argue that it is not helpful; this, however, is to beg the question for a moment. In an earlier paper, co-written with John B. Matthews, Jr., Goodpaster reframes the question of ‘usefulness’ in the following way: ‘As for holding corporations responsible, recent criminal prosecutions such as the case of Ford Motor Company and its Pinto gas tanks suggest that society finds the idea both intelligible and useful’ (1982: 139).

The passage from Goodpaster and Matthews, Jr. is helpful, inasmuch as it calls our attention to the major reason some find it useful to treat the corporation as if it were a moral person: the issue of assigning blame. One last way to assert that it still might be useful to treat corporations ‘as if’ they were moral persons might be to say that, even recognizing that a corporation is not a moral person, we still might treat them as if they had moral responsibilities. Rita Manning (1984: 82) seems to argue in much this way when she argues that if the ‘decision-making procedures and information gathering networks’ of the corporation are (even in part) responsible for bad things happening in the world, that we are correct to morally blame not just the individuals within the corporation, but the corporate institution itself. Manning’s argument, bizarrely, seems to run roughly thus: because corporations are not moral persons, we need not worry about whether or not we respect their moral rights (the right, say, to not be unfairly accused of moral wrongdoing). Therefore, ‘we do not need to show that our corporation is a person before we can show that it is morally at fault’, because it can never be ‘unfair’ to blame it anyway! This argument, of course, would seem to apply in much the same way to hurricanes; we need not worry about stepping on a hurricane’s moral rights, and so we are not being unfair to morally blame them for the damage they cause. But of course Manning isn’t interested in whether or not it makes sense to morally blame corporations – her concern is rather with the possible effects of assigning moral blame in this way. Manning writes that we assign moral blame ‘when we object to certain kinds of behavior which we want to discourage’ (ibid.). ‘We want to modify their behavior if we think it is inappropriate’, Manning continues. ‘In deciding how to do this, the considerations are utilitarian; we want to gauge the effectiveness of alternative courses of action’ (ibid.: 83). To say that it is useful to treat a corporation ‘as if’ it were a moral agent really comes down to saying that, even if a corporation is not a moral agent, it may still be useful to morally blame them.

But if a corporation is not a moral entity, then why should anybody think it useful to attempt to assign moral blame to one? To answer this question, allow me to briefly sketch out a little history. As many business ethicists have noted (e.g., Rowland, 2006; Banerjee, 2007; Painter-Morland, 2011), the granting of corporate charters used to be strictly tied to service of the public good. If and when a corporation’s actions violated this public good, it was once standard practice to revoke the corporate charter. However, over the course of the first half of the nineteenth century, this practice greatly dwindled, and by the twentieth century the restriction of corporate charters to service of the public good had disappeared. Today, as Mollie Painter-Morland has noted, ‘[c]orporations are no longer officially required to serve the public interest, and even though some laws govern their relationships with stakeholders, the law also grants them many rights and freedoms’ (2011: 18). As legal accountability has dwindled, calls for ‘business ethics’ have been on the rise. To venture a hypothesis (which it exceeds the scope of this paper to fully substantiate), I would suggest that the idea that it is useful to appeal to the corporation as a moral entity has arisen (at least partially) on the basis of our loss of the ability to restrict it legally. Those who think it is useful to treat the corporation as a moral entity (even though, as we have shown, it is not) seem to be appealing to a sort of exasperated modus tollens: ‘Well, if not this – then what?’

If my hypothesis seems overly speculative (or cynical), note how often the converse claim is made. ‘Only when they [corporations] acknowledge their responsibilities as secondary moral agents will corporations be able to carry out their obligations independent of coercive regulation’, writes Patricia Werhane (1985: 76). Likewise, Freeman and Reed explicitly frame their pitch for the stakeholder approach to CSR by suggesting that ‘a volunteeristic approach to questions of corporate governance which focuses on effective director behavior is preferable to structural change via legislation’ (1983: 88). Norman Bowie and Ronald Duska, meanwhile, warn that ‘business needs to police itself or be policed, so that limitations on self-interested profit are carried out where appropriate’ (1990: 95-96), and to this end explicitly recommend industry-wide codes of ethics as a way to help stave off government regulation. We have likewise seen business managers move quickly to adopt statements of corporate responsibility and business ethics training seminars as a PR response to corporate scandals. Is it such a stretch to imagine that the persistent myth that it might be useful to treat corporations as if they were moral agents emerges from precisely the failure to control them in more logically coherent ways (i.e. at the legal and economic levels)?

The objection here – alluded to, once again, by the reference Goodpaster and Matthews, Jr. make to the Ford Pinto case – will surely be that the by now thoroughly-entrenched idea of corporate legal personhood ought to be taken to mean precisely the opposite of what I am claiming. That is, corporate legal personhood means that we can place legal blame on corporations – so why should moral blame be thought of as some sort of supplementary substitute? In the Ford Pinto case, of course, Ford Motor Company was acquitted of charges of manslaughter; this actually makes it easier to gloss over some of the difficulties involved in trying a corporation for a crime. If we turn to a different case, however – for example, Granite Construction Co. v. Superior Court (People), a case in which the corporation charged was also found guilty – we might see the issues that arise when the court is forced to address the issues head-on.[7] In his ruling, Judge Harry Woolpert writes that the court has been ‘asked to exempt corporations from prosecution for manslaughter. We refuse, holding that corporations may be prosecuted for manslaughter under existing California law’ (Granite Construction Co. v Superior Court (People), 1983: 465). The interesting part of Woolpert’s ruling, for our purposes, is where it directly addresses the extent to which we can hold corporations responsible. When a human person is found guilty of a crime in the United States, there are a variety of punishments that can be doled out – ranging from community service, court-mandated therapy or classes, and fines, all the way up to incarceration and even (in some states) death. Judge Woolpert notes that, when it comes to ‘punishing’ a corporation, the only penalty that can actually be administered is a fine. The court must therefore impose ‘fines on corporations where both fines and imprisonment would be imposed on natural persons’ (ibid.: 471 n. 2). John Danley puts the issue a little more forcefully when he writes:

The corporation cannot be kicked, whipped, imprisoned, or hanged by the neck until dead. Only individuals of the corporation can be punished. What of punishment through the pocketbook, or extracting compensation for a corporate act? Here too, the corporation is not punished, and does not pay the compensation. Usually one punishes the stockholders who in the present corporate climate have virtually no control over corporate actions. Or, if the corporation can pass on the cost of fiscal punishment or compensation, it is in the end the consumer who pays for the punishment or compensation. If severe enough, hitting the pocketbook may result in the reduction of workforce, again resting the burden on those least deserving, more precisely, on those not responsible at all. (1980: 146)

Woolpert freely acknowledges the ‘inadequacy of the penalty provided by’ the law – but maintains that this ‘is a legislative problem irrelevant to this case’ (Granite Construction Co. v Superior Court (People), 1983: 471 n. 2). He offers that ‘the corporation [still] has reason to defend itself against the charge, because of the damage to its reputation, the standing of management in the eyes of its stockholders, and the like’. But when the stakes are corporate reputation and financial penalties (which are easily passed along to more innocent parties), then ‘holding corporations responsible for wrongdoing’ too often ends up simply becoming a challenge to PR firms and cost-cutting business consultants.

All of this is of course to say that it could be possible to hold corporations legally responsible for wrongdoings. Again, there was a time when such practice was quite standard, and when the granting of corporate charters was strictly tied to upholding the public interest. The current absence of such practices makes the turn to a certain kind of ‘business ethics’ – attempting to appeal to corporations as if they were moral agents – understandable, but no less unintelligible. I shall return to this point below, and I shall argue that notions of ‘corporate social responsibility’ only make sense as political goals. In other words, rather than attempting in vain to hold corporations morally responsible for their wrongdoings, we should instead refocus our efforts on broadening our ability to hold them legally responsible. For now, though, let us simply return to the point at hand and say: we can see why some might think it is useful to treat a corporation as if it were a moral agent. But the fact that the corporation is not a moral person ultimately undercuts any and all such usefulness.


It is at this point that we should return to corporate management and agency theory. For the valuable insight at the root of agency theory is that ‘business’ is a set of institutions that create certain kinds of agency – and these types of agency, while not bringing with them any specific moral responsibilities, do yet bring certain obligations, and provide a framework within which certain kinds of action can (and must) be taken. The exclusive focus within ‘agency theory’ seems to have heretofore been on the relationship between agents and principals – but the important thing to note here is that an agent is not simply one who acts ‘on behalf of’ another, but is also an individual granted a certain kind of agency. This agency brings with it certain abilities and also certain obligations. What obligations, then, does business agency carry with it? To answer this question we must place the business roles in question back within their context inside certain kinds of business institutions. Peter French (1977: 576) draws on the work of John Searle for this purpose:

An institution’s set of rules… is not merely a collection of devices that regulate antecedently existing behavior patterns. Rather the rules define, or, in Searle’s terms, are constitutive of, new kinds of activity; that is, they identify the performance of certain actions as ‘counting as’ the performance of an institutional act (e.g., from Searle, ‘A touchdown is scored when a player crosses the opponent’s goal line in possession of the ball while play is in progress’).

Taking on a role within a business thus grants a certain kind of agency, through the definition of that role within the larger institutional structure. A corporation, as an institution (composed of certain rules and roles, and granted a certain kind of legal status), will be designed to perform a certain function; this will most often be the function of generating profit, though Milton Friedman also cites the example of businesses designed with ‘an eleemosynary purpose – for example, a hospital or a school’ (1970: 33). A business agent thus gains certain abilities (to have his or her actions ‘count as’ certain kinds of legitimate business activities) to use in the fulfillment of certain responsibilities. Elaborating upon our earlier discussion of agents acting on behalf of corporations, Friedman tells us that the business agent takes on a ‘direct responsibility to his employers. That responsibility is to conduct the business in accordance with their desires, which generally will be to make as much money as possible while conforming to the basic rules of the society’ (ibid.). The point is not that a business agent must focus on profit, however (again, exceptions like schools and hospitals might be cited); rather it is that, outside of this institutionalized responsibility, it is essentially meaningless to talk about the ‘responsibilities’ of a business agent. This means, furthermore, that the ‘responsibilities’ that a business agent has qua business agent will only be the things she is required by her job to do, as French (1977: 577) explains:

Hence, the sentence… ‘X has an obligation to do something’, is a description of X’s institutional situation. It should be understood as equivalent to… ‘X is required by the rules of some institution to do something’.

It thus makes every kind of difference in the world just what kind of institution the practical role in question is embedded within, and how these institutions are designed.

It is tempting to try and use agency theory to bring business ethics back into the picture; the all-too-common approach here is to attempt to derive corporate moral responsibilities from an ‘implicit contract’ between a business and the society in which it is run, casting the society as another kind of ‘principal’. If one can show that ‘business’ as such has certain moral responsibilities to the community or society, then the agents who take on the responsibility of acting for and through businesses would thus inherit these special moral responsibilities. The advantage of framing the issue in this way is that, unlike the attempt to turn a corporation into a moral agent itself, it is entirely conceivable that an institution of any sort could be set up on certain conditions, and that these conditions might include the taking on of certain additional moral duties by the members of that institution. A fraternity or sorority, for example, might require its members to perform a certain number of hours of community service, and this requirement might be tied to the organization’s official recognition by either the school or the organization’s own central branch.

If you will recall the history I sketched out in section four, you will note that the social responsibility of business has during much of the history of corporations rested upon an explicit contract – and so the ‘role-specific’ responsibilities carried by agents of these corporations were not special moral responsibilities, but were legal responsibilities. And we should see once again that the appeal to ‘corporate social responsibility’ – the ‘implicit contract’ – is an attempted ad hoc solution to an age in which the legal responsibilities of corporations have all but disappeared. Unfortunately, with few legal, formal, explicit responsibilities built into business institutions, business ethicists have to appeal to an ‘implicit contract’ (see, for example, Carroll, 2008: 93; Bowie and Duska, 1990: 77). But what kind of a ‘contract’ is this – and, more importantly, what is its normative foundation? The idea of an ‘implicit contract’ seems to be consistently built upon references to ‘what society expects’ from business. How such expectations gain any normative traction, however, is unclear; it is certainly the case that I can expect things from you without you being in any way obligated to fulfill those expectations. Without any kind of explicit contract laying out ‘social demands’ on corporations, it’s unclear what kind of normative basis appeals to ‘implicit contracts’ carry – that is, without simply recurring to some theory of the corporation as a moral agent. This is why Patricia Werhane, though largely sympathetic to social contract theory, is yet led to caution us, ‘The theory assumes that corporations are moral agents, but it is less clear as to what this agency entails’ (1985: 46).

The obvious move, of course, is to say that the social and moral responsibilities of business are derived from the social and moral responsibilities of the people who own or run the business. In this sense, one looks not so much to derive moral responsibilities from agency theory, but rather to show how moral responsibilities direct, shape, or constrain that agency. Samuel Mansell (2013), for example, attempts to show that CSR is compatible with ‘orthodox shareholder theory’ by attempting to show that the desires (and supererogatory moral motives) of shareholders can obligate executives to pursue socially responsible ends. Freeman and Phillips (2002: 338) likewise argue that a business ‘is a nexus of contracts or the centerpiece of an ongoing multilateral agreement, based on voluntary consent…. If there is a weak presumption that the agreement is ongoing, managers must take the interests of all parties to the contract, or the nexus, into account’. In arguments of this type, the logic would seem to be something like this: a corporation is just a series of contracts – or promises – between people, and therefore the usual moral rules regulating voluntary agreement and promises between people apply to actions within the corporation. As Werhane puts it, this approach seems appealing ‘because it identifies any so-called “corporate moral agency” with the sum of the individual moral views of its constituents and thereby avoids some of the philosophical difficulties inherent in describing a corporation as a unit like a moral person’ (1985: 41). However, the problem – as Werhane (1985), Ciepley (2013), and Blair and Stout (1999) have all convincingly argued – is that the corporation is not simply a ‘nexus of contracts’, and the shareholders are not simply owners of property who can dispose of their property as they wish. Rooting agency theory in either a principal-agent relationship between the shareholder(s) and the executive(s), or a contractual relationship between various ‘stakeholders’ involved in the corporation, is to overlook the distinctiveness of the corporate form. And that distinct form, as it turns out, specifically foils the attempt to root business moral agency in the moral agency of the humans who own or operate a corporation.

A corporation, as John Ladd (1970: 488) reminds us, is a formal organization. Formal organizations ‘make a clear-cut distinction between the acts and relationships of individuals in their official capacity within the organization and in their private capacity’. The bureaucratic structures of formal organizations provide formal roles with prescribed duties, and individuals ‘take up’ these roles while remaining distinct from them. As a ‘decision-making structure’ (Ladd, 1970: 492), a formal organization is a rational structure for the pursuit of some pre-given end (in most cases, the pursuit of profit). Decisions of the organization – made through a constitution or ‘CID structure’ – are held up to the efficiency standard of rationality, and a ‘good’ or ‘bad’ decision is made within the distinct ‘language game’ of the organization:

The game not only determines what should and should not be done, but also sets forth the goals and the moves by which they are to be attained. More important even than these, a particular language-game determines how the activities within it are to be conceptualized, prescribed, justified and evaluated. (Ladd, 1970: 491)

The function of the bureaucracy is thus to split off the moral, private individual from the rational official. Decisions are ‘depersonalized’, decisions are made on behalf of the organization, and values and goals external or foreign to the goals of the organization are screened out, ‘automatically excluded as irrelevant to the organizational decision-making process’ (Ladd, 1970: 496). Patricia Werhane (1985: 43) captures the logic of the process quite nicely when she summarizes: ‘Therefore, while corporate activities are rule-governed, these rules, as impersonal operating procedures, preclude rather than imply moral agency’.

Of course, Werhane objects that this account of a corporation as a formal, decision-making structure – both Ladd (1970) and Danley (1980) describe it as a ‘machine’ – cannot account for the ways in which corporations respond to moral social pressures, or the ways in which people inside the corporation can sometimes make morally-motivated decisions. But on the contrary, Ladd’s account is quite clear. Regarding the ways in which individuals inside a corporation can sometimes make morally-motivated decisions, Ladd (1970: 490-1) points out that rationality functions within a formal organization as a normative standard – it is, he says, quoting Herbert Simon, ‘not a description of how administrators decide so much as a description of how good administrators decide’. Ladd studies the structure of formal organizations as a way of coming to understand a kind of wide-spread alienation and ‘moral schizophrenia’ he sees within modern societies; and this is because the standards for ‘morally good people’ do not shape or limit the standards for ‘good executives’, but often rather exist at loggerheads with them. The actions and decisions of a corporate executive are ‘subject to the standard of rational efficiency (utility)’, whereas the decisions and actions of ‘the individual as such are subject to the ordinary standards of morality’, and these two are, ‘at times, incompatible standards’ (Ladd, 1970: 501). Because corporate executives must also be individuals (that is, humans, moral beings), but cannot do both well or correctly at the same time, an inevitable moral schizophrenia results: the same person, trying to be live by two competing value systems at the same time.

Regarding the fact that corporations can and do react to moral social pressures, however, Ladd’s remarks are even more interesting. For while Werhane seems to believe that this is a sign that corporations can have a conscience, Ladd is right (I think) in pointing out that it is not the fact that they are moral pressures, but the fact that they are social pressures that makes corporations respond. Moral considerations, writes Ladd, can only ‘be relevant to the operations of a formal organization… by becoming limiting operating conditions’, that is, ‘conditions that set the upper limits to an organization’s operations’ (1970: 498).

It follows that the only way to make the rights and interests of individuals or the people logically relevant to organizational decision-making is to convert them into pressures of one sort or another, e.g. to bring the pressure of law or of public opinion to bear on organizations. (ibid.: 508)

It is simply a category mistake to appeal to corporations on the basis of ethical principles; but corporations can and will respond to coercion through law (and, under certain circumstances, organized social pressure). And where external, limiting conditions set the rules for a corporation’s actions, these rules will also govern the way managers and executives function – not just the ‘bad’ executives, who do their job ‘irrationally’ by making moral decisions, but also and especially the ‘good’ executives, who decide rationally and efficiently.

While the attempt to use agency theory as a way of normatively grounding business ethics is thus in error, this very failure provides us with an important insight. The creation of business agency does bring with it certain obligations, but these obligations can only be tied to explicit contracts – this is why agency theory has focused almost exclusively on the relationship between agents and principals. Business agency is shaped by the business structures – formal institutions – in which they are embedded, however; change the institutional structure, and you change the form of agency, as well as its attendant responsibilities. If we want to establish extra-economic responsibilities for business agents, then, we need to change the institutions rather than attempting to appeal to the moral nature of these agents. And such changes will not take the form of appealing to the consciences of corporations, but of changing their structure. The failure to ground business ethics in agency theory, then, should not be taken as a problem with agency theory – but as a sign that we are mistaken to ground CSR in business ethics.


I have now argued that business is not – and should not be seen as – a moral entity, bringing with it any specific moral obligations (or exemptions); and that corporations are not moral subjects, and hence cannot – and should not – be beholden to moral obligations. Furthermore, I have argued that much of the intuitive appeal of seeking to morally praise or blame the actions of corporations comes from the lack of legal and political recourse. If CSR theorists are increasingly turning to ethics as an ad hoc solution for the loss of any kind of legal mechanisms of business accountability, Bowie and Duska are quite right to warn us that such an approach is insufficient for the task:

[A] single individual or a single firm alone will find it next to impossible, unless they are heroic, to overcome or counteract the generally accepted practices of business, even where those are unethical. Thus, business needs to police itself or be policed, so that limitations on self-interested profit are carried out where appropriate. (1990: 95)

Furthermore, they add, even when businesses or individuals are ‘heroic’, morally heroic decisions – under unregulated market conditions – are liable simply to put one out of business (ibid.: 96). Picking up on the refrain that is so common in business ethics literature, Bowie and Duska recommend industry-wide adoptions of moral codes as a way to stave off government regulation; but by the end of the book, the authors have made it clear that some kind of government regulation is probably necessary to counteract the built-in profit-seeking function of businesses (ibid.: 102).

At the heart of business ethics approaches to CSR, then, we have a failure married to a confusion. The failure comes from attempting to address morally problems which are better addressed politically. And the confusion is one of normative scope; when either ‘business’ or corporations are taken as the moral subject, then business ethics commits itself to this failure by attempting to derive specific moral guidelines and obligations where no such guidelines and obligations can exist. Given the limitations and failures of this sort, it seems to me that many of these problems would be better addressed by turning to something like business politics. If there is currently no such thing as ‘Corporate Social Responsibility’, it is only because social responsibilities are no longer required as a legal condition of incorporation. If there are no moral limits placed upon business agents qua business agents, it is only because industry-wide, externally-enforced codes of conduct have not been adopted. ‘Corporate Social Responsibility’ only makes sense, in fact, as a political goal. The practice of business and the granting of corporate charters have both in the past been limited by legally-instituted and -enforced social goals. Those who desire to see more ‘Corporate Social Responsibility’, then, would do better to focus their efforts on legal and political approaches to solving the problem, rather than appealing to baseless ethical notions of corporate moral agency.

Note, furthermore, that the distinction between ‘business ethics’ and ‘business politics’ completely dissolves the major objection to talk of ‘Corporate Social Responsibility’. As Milton Friedman (1970) has framed it, this objection roughly runs, ‘The only social responsibility of business is to increase profits’. As I have already argued above, I believe Friedman to be entirely correct when he argues that, qua business agent, a business agent’s responsibilities are derived from the specific kind of agency granted to him or her by participation in a business institution. Friedman writes:

In a free-enterprise, private-property system, a corporate executive is an employee of the owners of the business. He has direct responsibility to his employers. That responsibility is to conduct the business in accordance with their desires, which generally will be to make as much money as possible while conforming to the basic rules of the society, both those embodied in law and those embodied in ethical custom. (1970: 33, emphasis added)

While much time and effort has been spent quibbling over Friedman’s latter phrase, ‘embodied in ethical custom’, it should be clear that any rules regulating business ‘embodied in law’ are simply accepted by Friedman (and business) as legitimate constraints. When we transform ‘Corporate Social Responsibility’ from an ethical demand into a political goal, we sidestep almost completely the neoliberal objection that such talk provokes.[8]

Especially among those who desire to see more Corporate Social Responsibility, however, there is likely to be one more major objection to the position I am putting forward. The objection, I take it, would be something like this: I am drawing a distinction between ‘business ethics’ and ‘business politics’, where in fact no such distinction exists. All struggles for corporate social responsibility must be rooted in ethical principles – and therefore I am merely suggesting a better (and perhaps more political) strategy, while any normative claims (about what the role of business ought to be) must continue to take the form of ‘business ethics’. Without business ethics, in other words, there can be no business politics. My response to this objection is twofold. First, pragmatically speaking, it is simply not true that any idea of Corporate Social Responsibility must be rooted in ethics; indeed, if what I have been saying is correct, any such rooting is at best highly misleading. And second, even if we were to admit that discussions of ‘social responsibility’ are in some sense discussions of ‘moral principles’, it is yet the case that the attainment of these principles themselves are a political goal, rather than the kinds of things that can be derived from ethical theory; in other words, ‘business ethics’ as a discourse in itself must arise from, rather than ground, politics. In this final section of my essay, allow me to briefly explain my responses.

Thanks to the discourse of business ethics, it has come to be assumed that all ‘Corporate Social Responsibilities’ are moral responsibilities. But this needn’t at all be the case. Might we want to regulate corporations in order to demand more honesty, more respect of human beings, and more environmental responsibility from them? Certainly, we probably do. But could we also collectively – and through the force of law – demand other, far more pedestrian responsibilities of them? Could we demand that corporations aid the building and repair of infrastructure, support education, or even devote resources to litter removal or other community service? Yes, we could. My point is not that any of these goals might be immoral – only that these goals need not in any sense reflect the kind of moral responsibilities that normal moral agents are said to have. These responsibilities at the very least could be the kinds of things that would be supererogatory for normal moral agents – and could even take the form of desired community improvements that would go far beyond ideas of ‘charity’. (Furthermore, such activity would not be ‘philanthropic’, as it would be done from necessity rather than generosity.) If we were to grant corporate charters on the condition that the existence of a corporation also serves social goals, these ‘social goals’ could be anything, rather than merely the demand for ‘minimally-decent Samaritans’.

Furthermore, if what I have been saying is true, then the demands we place upon corporations could not take the form of demands for ‘morally responsible behavior’. Morally responsible behavior is the behavior of a moral agent who lives up to his or her moral responsibilities; as corporations are not moral agents, they do not have moral responsibilities for us to demand that they fulfill. Rather, as a piece of technology, a corporation exists to serve ends we have decided upon ahead of time. Rather than merely serving financial ends, I am arguing that we could return to the days of demanding that corporations serve other, socially desirable ends as well.

One might still object, however, that any discussion of which ends are ‘socially desirable’ must be a discussion of ethics, inasmuch as we are addressing issues of the Good. And so, even if it has the aim of being codified in law and governmental regulation, one might object that these discussions of ‘the social ends which business ought to serve’ would still be a form of ‘business ethics’, and that furthermore this business ethics would therefore still have to underwrite any ‘business politics’. While I accept the principle that moral judgment will inevitably be involved in each individual’s judgment of ‘the ends a corporation ought to serve’, I must yet reply that this objection seems to confuse the goal of such a discussion for its form. As Scherer and Palazzo (2007: 1104) have also argued, we do ‘need a normative discussion on how the legitimate role of business in society should be defined and how the business firm should act in a responsible way’, but such a discussion is the goal of business politics, not its precondition. In other words, the guiding principles of CSR are not conditions to be derived in advance from a theory of ethics, and then put into place through legal codification. Rather, such principles ought to be the result of democratic discussion; the arrived-at principles will be legitimate, not to the extent that they are certified by one theory of ethics or another, but by their having resulted from as widespread a democratic consensus as possible.[9] Another way of framing my demand for Corporate Social Responsibility is to say that I am looking to restore the democratic control of corporations – or at least of some of their necessary ends. Corporate governance ought to be determined through democratic process and political struggle, rather than through moral appeals to either business agents or lawmakers. It is not ‘business ethics’ underlying ‘business politics’, in other words, but exactly the opposite: business politics and the restructuring of corporate law will allow us to discuss ‘business ethics’ (among other socially-desirable goals!).

In summary, then, we will certainly want to continue to have normative discussions about the role corporations can play in society. But these prescribed ends need not be strictly moral, the normative content of the conversation need not restrict itself to strictly moral considerations, and the content of these judgments is to be arrived at democratically rather than derived from ethical theory. And so my critics and I may disagree about the extent to which ‘business ethics’ as a discourse might continue to have any meaning at all; but I must at least insist that my call for ‘business politics’ is not simply a pragmatic strategy for business-ethics-as-usual!

Richard Marens (2013) has argued (rather persuasively) that CSR as a discourse about corporate management has arisen and taken hold in the wake of the loss of explicit, legal forms of business regulation. In this sense, CSR has developed as a discourse of justification, a moral smoke screen to fend off either renewed regulations or public backlash at their repeal. If my argument thus far has been correct, then we can more or less agree with Marens: CSR as a moral discourse has arisen in response to an inability to control corporate behavior. But we needn’t throw the baby out with the bathwater. As Henry Mintzberg puts it, ‘Social responsibility – that most naïve of concepts – represents our best hope, perhaps our only real hope’ (2007: 50). Despite having spent his article showing why CSR is impossible (Mintzberg’s approach to CSR is rooted in the standard, business ethics approach), Minztberg yet argues that it is necessary. I have tried to argue that this failure of CSR (the tragedy, if you will, of an idea that is both necessary and impossible) derives from our tendency to approach it as a moral goal. If we instead approach CSR as a political goal, restructuring both business institutions and the legal frameworks in which they are situated, then ‘corporate governance’ will cease to be a competing paradigm, and become instead a set of tools for CSR’s implementation. There is obviously much difficult work to be done, here, and I do not begin to claim that my article has even scratched the surface of such difficulties. Rather, it is my hope that this essay might serve as a ‘ground-clearing operation’: by showing why business ethics approaches to CSR are fundamentally mistaken, it is my hope that we might begin, together, the difficult work of mapping out a business politics approach.


[1]       There is by now a wealth of diverse literature criticizing this dominant approach; but the standard introduction is to acknowledge stockholder theory and the principal-agent problem as the dominant approach before moving to criticize it. See, for example, Daily, Dalton, and Cannella (2003: 371), who begin by noting that agency theory is the ‘overwhelmingly dominant theoretical perspective applied in corporate governance studies’. Likewise, Rajan and Zingales (2000: 1) explicitly lay out their understanding of ‘the new enterprise’ by showing how things have changed since Berle and Means ‘set the terms of the modern debate on corporate governance’ with their work on the principal-agent problem. So too do Blair and Stout (1999: 248) begin their account of the ‘team production theory’ of the corporation by noting that ‘discussions of corporate governance have come to be dominated by the view that public corporations are little more than bundles of assets collectively owned by shareholders (principals) who hire directors and officers (agents) to manage those assets on their behalf’.

[2]       See, for example, Carroll (2008), the famous ‘pyramid of CSR’; see also Freeman and Reed (1983). R. Edward Freeman has of course made a number of principled arguments for stakeholder theory (e.g., Freeman and Phillips, 2002), but he is not above occasionally arguing that corporations will be more successful (read: profitable) if they treat their stakeholders well.

[3]       E.g., ‘Both the individual decision-maker within the organization and the organization itself as a decision-making entity will need our attention’ (Goodpaster, 2007: 15).

[4]       French’s assertion that corporate decision-making structures can be formally identified (and, hence, that a hard-and-fast distinction can be made between corporations and crowds or mobs) has been widely criticized (see, e.g., Danley, 1980) – as an ‘ideal’ process, it is very likely that the sort of decisions French thinks we can attribute to corporations rarely (if ever) happen, while the sorts of ‘corporate actions’ that we most want to morally censure are actions that do not (indeed, cannot) be said to follow such a path. (Indeed, some researchers even speak of ‘the “black box” of boardroom deliberations’: Daily et al., 2008: 379) Furthermore, as Christian List and Philip Pettit (2006) have shown, even at the level of formal constitution, not all ‘CID structures’ are created equal. A wide range of CID structures (List and Pettit simply refer to ‘constitutions’, a category which would seem to include – but not be limited to – the two-part CID structures French discusses), including ‘proposition-wise’ democracies, will result in irrational group agents, even if these groups are composed of fully rational individuals. In other words, ‘rationality’ will only supervene at the group level if it is specifically built into the constitution from the beginning. Even if CID structures were fully followed in the real world, then, this still would not be enough to always and automatically guarantee rational agency at the level of the corporation.

[5]       Peter French is quite right when he claims, early on in his argument, that ‘to be a moral person, the subject must be at minimum what I shall call a Davidsonian agent’ (1979: 211, emphasis added), where ‘Davidsonian agent’ is here equivalent to what we have been calling, in Dennett’s language, an ‘Intentional System’. But by the end of his article, French simply pretends that this minimal condition is enough: ‘I have maintained that Davidsonian agency is a necessary and sufficient condition of moral personhood’ (1979: 215). He repeats this trick in his later article ‘Kinds and persons’ (1988), again without any argument to support it; early on in the article he writes that ‘to be a person is, at least, to be an intentional agent’ (302; emphasis added), before simply declaring several pages later that intentional agency is both a necessary and sufficient condition of personhood (306). Nor, as some might claim, does French’s later (1984) elaboration of the principle of responsive adjustment (‘PRA’) do anything to alleviate the problem. Here he simply assumes that corporations are persons, and his question is rather about whether and to what extent persons can be held morally responsible for the harmful effects of non-intentional behavior (1984: 101).

[6]       My intention in this section is still to criticize a position, rather than simply to criticize specific theorists. However, I believe that Werhane and Copp, between them, have laid out this position about as clearly as one could hope for – and so I will tend to stick fairly closely to their texts as I discuss this position.

[7]       This ruling is available online at
 (accessed 1 October 2011).

[8]       Almost completely. Certainly, the neoliberal objection is bound to now run something like this: ‘Increased governmental regulation of business – including the conditions of incorporation – can only have the effect of making businesses less competitive, and therefore drive business away’. Objections like this have been addressed many times in the past – but the important point here is that such a debate at least properly reframes arguments about ‘Corporate Social Responsibility’. As those debates now stand, neoliberals can rightly accuse business ethicists of pernicious metaphysical nonsense; a charge business ethicists are left attempting to refute by arguing, as Goodpaster does, that such nonsense might yet be ‘useful’ – even as realists like Bowie and Duska recognize the inherent futility in such approaches.

[9]       To this extent, I agree with Scherer and Palazzo, and find myself sympathetic to their ‘Habermasian’ approach. However, as their 2011 article, ‘The new political role of business in a globalized world’ shows, Scherer and Palazzo have come to take up a position that is not at all what I mean by ‘business politics’. In their recent work, Scherer and Palazzo seem to want to integrate corporate institutions (albeit democratically controlled corporations) with the democratic institutions of the state. Rather than regulating corporations through state power, Scherer and Palazzo seem to want corporations to co-govern! A criticism of their position is of course far beyond the scope of this essay – but suffice it to say, I am not convinced.


Arnold, D.G. (2006) ‘Corporate moral agency’, Midwest Studies in Philosophy, 30: 279-291.

Banerjee, S.B. (2007) Corporate social responsibility: The good, the bad, the ugly. Northampton, MA: Edward Elgar.

Blair, M.M. and L.A Stout (1999) ‘A team production theory of corporate law’, Virginia Law Review, 85(2): 247-328.

Bowie, N.E. and R. Duska (1990) Business ethics, 2nd ed. Englewood Cliffs, NJ: Prentice Hall.

Carroll, A.B. (2008) ‘The pyramid of corporate social responsibility: Toward the moral management of organizational stakeholders’, in A. Crane, et al. (eds.) The Oxford Handbook of Corporate Social Responsibility. Oxford: The Oxford University Press.

Ciepley, D. (2013) ‘Beyond public and private: Toward a political theory of the corporation’, American Political Science Review, 107(1): 139-158.

Copp, D. (1979) ‘Collective actions and secondary actions’, American Philosophical Quarterly, 16(3): 177-186.

Corvino, J. (2006) ‘Reframing “morality pays”: Toward a better answer to “why be moral?” in business’, Journal of Business Ethics, 67: 1-14.

Daily, C.M., D.R. Dalton, and A.A. Cannella, Jr. (2003) ‘Corporate governance: Decades of dialogue and data’, Academy of Management Review, 28(3): 371-382.

Danley, J.R. (1980) ‘Corporate moral agency: The case for anthropological bigotry’, in M. Bradie and M. Brand (eds.), Action and responsibility. Bowling Green, OH: Bowling Green State University.

Dennett, D. (1988) ‘Conditions of personhood’, in M.F. Goodman (ed.) What is a person? Clifton, NJ: Humana Press.

Freeman, R.E. (1994) ‘The politics of stakeholder theory: Some future directions’, Business Ethics Quarterly, 4(4): 409-421.

Freeman, R.E. and R.A. Phillips (2002) ‘Stakeholder theory: A libertarian defense’, Business Ethics Quarterly, 12(3): 331-349.

Freeman, R.E. and D.L. Reed (1983) ‘Stockholders and stakeholders: A new perspective on corporate governance’, California Management Review, 25(3): 88-106.

French, P.A. (1977) ‘Institutional and moral obligations (or merels and morals)’, The Journal of Philosophy, 74(10): 575-587.

French, P.A. (1979) ‘The corporation as a moral person’, American Philosophical Quarterly, 16(3): 207-215.

French, P.A. (1984) ‘The principle of responsive adjustment in corporate moral responsibility: The crash on Mount Erebus’, Journal of Business Ethics, 3(2): 101-111.

French, P.A. (1988) ‘Kinds and persons’, in M. F. Goodman (ed.) What is a person? Clifton, NJ: Humana Press.

Friedman, M. (1970) ‘The social responsibility of business is to increase its profits’, The New York Times Magazine, 13 September.

Granite Construction Co. v Superior Court (People) (1983) 149 Cal. App. 3d 465.

Goodpaster, K.E. (2007) Conscience and corporate culture. New York: Blackwell.

Goodpaster, K.E. and J.B. Matthews, Jr. (1982) ‘Can a corporation have a conscience?’ Harvard Business Review, 60: 132-141.

Hess, K.M. (2013) ‘“If you tickle us…”: How corporations can be moral agents without being persons’, Journal of Value Inquiry, 47: 319-335.

Ladd, J. (1970) ‘Morality and the ideal of rationality in formal organizations’, The Monist, 54(4): 488-516.

List, C. and P. Pettit (2006) ‘Group agency and supervenience’, The Southern Journal of Philosophy, 44: 85-105.

Manning, R. (1984) ‘Corporate responsibility and corporate personhood’, Journal of Business Ethics, 3: 77-84.

Mansell, S. (2013) ‘Shareholder theory and Kant’s “duty of beneficence”’, Journal of Business Ethics, 117: 583-599.

Marens, R. (2013) ‘What comes around: The early 20th century American roots of legitimating corporate social responsibility’, Organization 20(3): 454-476.

Mintzberg, H. (2007) ‘The case for corporate social responsibility’, in A. Crane, D. Matten, and L,J. Spence (eds.), Corporate social responsibility: Readings and cases in a global context. London: Routledge.

Painter-Morland, M. (2011) ‘Agency in corporations’, in M. Painter-Morland and R. ten Bos (eds.) Business ethics and continental philosophy. Cambridge: Cambridge University Press.

Rajan, R.G. and L. Zingales (2000) ‘The governance of the new enterprise’, National Bureau of Economic Research, working paper 7958.

Rorty, A.O. (1973) ‘Persons, policies, and bodies’, International Philosophical Quarterly, 13(1): 63-80.

Rowland, W. (2006) Greed, Inc.: Why corporations rule our world, and how we let it happen. New York: Arcade Publishing, Inc.

Scherer, A.G. and G. Palazzo (2007) ‘Toward a political conception of corporate responsibility: Business and society seen from a Habermasian perspective’, Academy of Management Review, 32(4): 1096-1120.

Scherer, A.G. and G. Palazzo (2011) ‘The new political role of business in a globalized world: A review of a new perspective on CSR and its implications for the firm, governance, and democracy’, Journal of Management Studies, 48(4): 899-931.

Werhane, P.H. (1985) Persons, rights, and corporations. Englewood Cliffs, NJ: Prentice Hall.

the author(s)  

Matthew Lampert completed his PhD in philosophy at the New School and is currently an adjunct professor of philosophy at several colleges in the New York City area. His recent and forthcoming publications include essays on critical pedagogy, Louis Althusser’s theory of ideological apparatuses, and the work of Jacques Rancière.
Email: lampm414 AT