The blame game: Defining the moral boundaries of organizations

The way that we perceive and interact with organizations often defies legal definition. For example, we often blame multinational enterprises for unethical behavior by one of their local subsidiaries, even though subsidiaries are legally separate entities. Does this mean that our instinct to ascribe blame in this way is misguided? Or is there something morally important to this instinct? From a moral perspective, where does the organization stop and the rest of the world begin? 

In 1937, the British economist, author and Nobel laureate Ronald Coase published an article that would come to shape to a great extent the way that we think about organizational boundaries. Coase viewed organizations as objects that people use to reduce the costs of economic cooperation. In his seminal article, he argued that organizations stop at the point where managerial direction is no longer more efficient than market exchange: that is, their boundaries are drawn so as to reduce transaction costs.  

An organization must be viewed not as an object that is used to accomplish a goal but as a subject in itself

Coase’s analysis of organizational boundaries, and his view of organizations as objects that are used to increase efficiency, has influenced decades of scholarly work, including much of the management literature. But this approach does not help us to answer questions about moral responsibility: it is intended for something quite different.  

New research from Rita Mota (Esade) and Alan D. Morrison (Saïd Business School, University of Oxford) aims to reframe the discussion on organizational boundaries in explicitly moral terms, and to shed light upon difficult questions about how to assign moral responsibility in complex settings. To do so, they argue, an organization must be viewed not as an object that is used to accomplish a goal — but as a subject in itself. In other words, the organization must be viewed as a moral agent.   

The organization as an agent

The question of whether organizations can be held morally responsible for their actions, and, so, if they can possess moral agency, is foundational in business ethics; it is also incredibly complex. 

There are essentially two different ways to think about corporatmoe moral agency. Under an individualistic approach, we start by asking what internal characteristics an organization must have in order to qualify as a moral agent; then, if those characteristics are present, we can say that it makes sense to hold the organization morally responsible. As a result, people’s emotional responses to organizational actions, like praise and blame, are appropriately justified. 

Under a relational approach, we start the analysis the other way around: we first ask if people have emotional responses to organizational actions and, if they do, we investigate if they persist when we scrutinize the internal characteristics of the organization. When those responses survive scrutiny, the organization is a moral agent. 

This research offers a process that provides clarity to the moral responsibility of organizations, their networks, and the individuals who work within them

Mota and Morrison suggest that a relational approach is more appropriate to investigate questions about moral boundaries. In particular, they argue that the corporate second-personal (CSP) account of corporate moral agency, which they developed in an earlier paper, is better suited than others: it is the most closely aligned with the social processes by which we hold collectives responsible, it provides a straightforward list of criteria, and it ensures the possibility of engaging in moral dialogue with the organization. 

Under the CSP account, an organization is a moral agent if it satisfies three conditions: first, it must be possible for people to form reactive attitudes (like praise or blame) toward it; second, it must be possible to trigger a dialogue with the organization by addressing reactive attitudes to it; and, third, the organization must be capable of acknowledging the moral authority of other agents and, hence, of responding appropriately when it receives a reactive attitude.  

According to Mota and Morrison, only moral agents can have a moral boundary; and that boundary encompasses the actions that can meaningfully be attributed to that agent – both the actions that it has performed and those that it can in principle perform. Because it is impossibly difficult to list every single potential action that an organization could perform, it is impossible to perceive an organization’s moral boundary in its entirety. So, the important question is: given a specific organizational action, does it fall within the organization’s moral boundary or not?  

Six steps to moral attribution

The researchers have developed a process for determining whether or not a given action lies within the moral boundary of an organization. In other words, they have developed a model that allows us morally to attribute an action to the right agent, and, hence, to hold the right agent morally responsible for that action. 

1. Identify the enactors. Who performed the action? This first stage requires a candidate organization (CO) to be identified, with the possibility of further identifying one or more individuals who are part of that CO. More than one CO can exist: joint moral responsibility is possible, and it may be necessary to go through the steps separately for several agents. 
2. Characterize the organizational action (OA). Is this action meaningfully organizational? The answer is yes if its human enactors are authorized, instructed, or expected to take the action on behalf of the organization; in this case , the process moves to step three. If the answer is no — for example, because a ‘lone wolf’ acted against the interests of the organization — the process moves to step six. 
3. Moral agency. For the CO to have moral boundaries, it must be a moral agent. Step three is about assessing whether this is the case, and the authors argue that the CSP account provides the appropriate test. If moral agency is confirmed, the process moves to step four. If it is not present, stage five offers clarity. 
4. OA attributable to CO. Quite simply, if the CO is a moral agent, then OA lies within its moral boundary. This means that OA is morally attributable to CO, and CO can be held morally responsible for it. 
5. Controlling organization. If the CO is found not to be a moral agent, then OA could lie within the moral boundary of a controlling organization (that is, an organization with formal or informal ownership and control of the CO). If such a relationship is present and a new CO is identified, the process is repeated from step three. If no controlling organization is present, the process moves to step six. 
6. OA attributable to human actors. In the event that no organization can be held morally responsible for the OA, the human actors responsible for the action must be identified. In this case, the moral responsibility for the action is “assigned downwards” to an individual rather than to the organization. 

The power to create, destroy, and change boundaries

The moral boundaries of an organization can be created, destroyed, or changed as a result of the choices made either inside or outside the organization. 

The creation of a moral boundary happens when an organization becomes a moral agent. The authors argue that the organization itself could make this choice, but only if it already has some ability to deliberate and to set its own goals. For example, a secret society may already have deliberative capacity and the ability to recognize others’ moral authority, but, because it is invisible to the world, it is not a moral agent by Mota and Morrison’s account. It can, however, choose to make itself visible to the world and to ensure that it receives, processes, and responds to reactive attitudes appropriately. If it does so, it becomes a moral agent and, so, it creates its own moral boundary. 

Moral boundaries are destroyed if any of the conditions for moral agency ceases to be satisfied. This can happen, for example, if, over time, a CEO becomes entrenched and, because of their unusual prominence and control, outsiders start forming reactive attitudes towards the CEO rather than the organization. In this case, it is also possible that the CEO controls any dialogue with outsiders so completely that it no longer makes sense to talk of organizational responses. Consequently, the organization loses moral agency and, hence, its moral boundary is destroyed. 

Finally, organizations can change their moral boundaries by altering the range of actions that they can perform. This can happen quite quickly through change in personnel or structure.  

The complexities of corporate moral responsibility

Many of the organizations that we interact with daily are part of incredibly complex networks that involve multiple other organizations. In some cases, if we attributed moral responsibility in the same way that we attribute legal responsibility to organizations, we would be making a mistake, and we would probably also be unfair. But the attribution of moral responsibility is not a straightforward task and, so far, we had very few tools to help us in that task. 

Mota and Morrison shed light on this matter by encouraging us to step away from the traditional Coasean perspective on organizations: rather, the authors compel us to view organizations as subjects in their own right, who can act in the world. By viewing organizations as subjects that act, rather than passive objects that exist, the focus can be turned to the responsibility of the agent. Its deliberate choices can then be identified, and it can be held to account for those choices. 

This research by Mota and Morrison not only recasts the definition of organizational boundaries, but also offers a framework to identify the actions for which an organization can be held morally responsible.  

This analytical framework offers, for the first time, a process that provides clarity to the moral responsibility of organizations, their networks, and the individuals who work within them.