The Problem with the Traditional Language of Corporate Purpose

Despite decades of debate and thousands of law review articles, no consensus has emerged around the fundamental question of what the corporation’s purpose is or should be. One likely reason for this lack of progress is a disagreement between different views of political economy, which are often tenaciously held. Another reason is that strong empirical evidence in support of this or that theory of corporate purpose is difficult to find. But strong policy disagreement and unpersuasive empirical evidence are ubiquitous in academic disputes, and yet very few disputes have the stubborn persistence and seeming inconclusiveness of the corporate purpose debate. These reasons cannot be the whole story.

In a new paper, forthcoming in the Yale Journal on Regulation, I argue that one of the major problems afflicting the debate is the use of a language that treats “corporate purpose” as a primitive concept rather than shorthand for a complex web of legal relations. Viewed as a primitive concept, corporate purpose becomes a coherent “thing” that can be oriented or re-oriented towards this or that normative goal. In this vein, the traditional characterization of the debate is a clash between two “contested visions”—shareholder primacy vs stakeholder governance. Should corporations “be managed… to maximize shareholder wealth”?  Or should they “prioritize the interests of other stakeholders”?

This traditional way of framing the problem requires that one takes a position about which orientation “corporate purpose” should have, which leads to vague or ambiguous claims. For example, the “rise of shareholder primacy” is commonly said to have occurred in the 1970s or 1980s, yet the idea that corporations must make profits for shareholders appears in law casebooks as early as the end of the nineteenth century.  The concept of “stakeholders” has been borrowed from the management literature, but management scholars have proposed 55 different definitions of this term over four decades. The term “ESG”—despite the explosive growth of funds using this label —lacks definitional coherence. Even the very concept of “corporate purpose,” once used in a technical legal sense, is now increasingly vague and fluid.

The thesis advanced in my paper is that corporate purpose, just like other central concepts in the debate—such as “shareholder primacy” or “stakeholder governance”—is what Karl Llewellyn called a “lump concept,” an overbroad label which obscures its more basic components. The ensuing confusion is vividly illustrated by a curious fact. Under the banner of “stakeholder governance,” we can find both Bernie Sanders, a self-styled democratic socialist, and the Business Roundtable, a powerful business lobby.  Something is amiss.

To have a more meaningful conversation, we need to shift the analysis from these lump concepts to their atomistic components. Beneath the surface of concepts like “purpose,” I argue, there is an interlaced matrix of legal relations. The real disagreement lies in the specific design of this matrix not in the vague holistic properties of the lump concept. But to talk about this legal matrix we need a sharper language—and a common one, so that we can constructively compare, assess, and discuss competing proposals. I argue that the work of Wesley Hohfeld provides the foundations of such a language.

Wesley Hohfeld’s Toolbox

Hohfeld—an early twentieth century legal philosopher well known among legal theorists but obscure to most corporate law scholars—argued that the corporation is a mere fiction and that all legal concepts can be reduced to a few basic relations between individual actors with respect to specific activities. I use his framework to propose a sharper, clearer language to talk about corporate purpose.

Just like Hohfeld showed that the traditional concept of “property”—among others—hides a bundle of different interpersonal legal relations, I hope to show that familiar concepts used in the corporate purpose debate—including the notion of corporate purpose itself—are in fact complex bundles of legal relations among market players and the state. Similarly, just like Hohfeld showed that what we commonly call “rights” and “duties” can only be understood in relation to each other—in other words, that there is no “right” without a correlative “duty” and vice versa—I hope to show that stakeholder-oriented corporate governance cannot rely on entitlements without corresponding disablements, or on disablements without corresponding entitlements.

The Hohfeldian architecture of corporate purpose is adversarial in nature, meaning that the position of each actor can only be understood in contraposition to other actors limiting and constraining the former. This implies that we cannot talk of duties, obligations, purposes, or missions—as in the pervasive rhetoric of corporate purpose—without identifying adversarial rights, powers, and other entitlements that give full meaning and operational efficacy to those concepts.

The Hohfeldian Architecture of Corporate Purpose

Consider a simple business decision: Should the company build a highly polluting plant or a more expensive, greener facility? For Hohfeld, a legal relation is always a relation between two actors with respect to one activity. Therefore, once we identify the activity (building the “green” facility), we must figure out who are the actors sitting at the two ends of a specific relation (an entitlement and a correlative disablement) with respect to that activity.

Framed in these terms, a Hohfeldian model of the business decision to build the “green” plant can take several different forms, as follows:

In Model I, shareholders have the right that the corporate decisionmaker shall not build the green plant at the expense of the shareholders and, therefore, the CEO has a correlative duty not to. In Model II, the CEO may choose to build the green plant but have no legal obligation to (Hohfeld calls this entitlement a “privilege.”) In Models III, IV and V, the CEO has a duty to build the green plant, but the actor holding the correlative right is different in each specification: the shareholders in Model III, the state in Model IV, and the interested stakeholders in Model V.

These five models represent concrete configurations of legal relations with respect to a very specific choice between two incompatible options—building a green plant or a more polluting plant—of which one is better for stakeholders and the other is better for shareholders. A very tempting analogical move is to generalize from these specific relations to fundamental conceptions of corporate purpose.

Indeed, the main theories of corporate purpose that are often discussed in the literature bear a very strong resemblance to these five permutations. Model I, for example, correspond to a hard version of shareholder value maximization, in which managers have an actual Hohfeldian duty to choose the value-maximizing option. Model II would be a permissive theory of managerial discretion, in which managers may choose the stakeholder-favoring option but do not have to. And so forth.

However, in real-world corporate decision-making, we find different types of relations—whether Model I, or Model IV, or Model II—in different types of situations. None of these relations, however, imply a general principle of the same form. As I try to demonstrate in the paper, there is not, and there could not practically be, a general Hohfeldian duty to always choose the value-maximizing option or the socially responsible option.

Any realistic model of corporate decision-making must take the form of a broad sphere of managerial privilege constrained by adversarial legal relations—rights and powers of shareholders, stakeholders, and the state that limit and shape how managers use their legal privilege. Therefore, the crux of the problem is not in the abstract commitment to shareholders or stakeholders, but in the concrete scope of managerial privilege, and how it is constrained by adversarial legal relations.

Proposals for corporate purpose reform cannot be as simple as a “re-orientation” of purpose from one goal to another.  Instead, they must be detailed and explicit specifications of which new or different legal relations should be established, between which actors, and based on what assumptions about the reasons and motives of corporate actors.

Reframing the Corporate Purpose Debate

Rereading the history of the corporate purpose debate through the lens of Hohfeld’s scheme reveals a major fault line since the 1930s: not the abstract contraposition between shareholderism and stakeholderism, but one between models seeking to expand managers’ privilege and models seeking to restrict it. Interestingly, whether someone proposes an expansion or restriction of managerial privilege often depends on underlying assumptions on the benevolence of corporate actors.

Merrick Dodd, for example, influenced by Louis Brandeis, thought that corporate managers believed in social responsibility and therefore an expansion of managerial privilege would lead to better social outcomes. Thirty years later, Milton Friedman shared the same assumptions on the sociology of managerial power, but given his very different policy preferences (Dodd was a New Deal corporatist, Friedman a free-market libertarian) Friedman drew the opposite conclusion—that managerial privilege should be restricted by value-maximizing duties.

Today, many influential “stakeholder governance” proposals rely on a similar bet—often understated or unspoken—on the benevolence of corporate actors. The bet, however, is quite risky, because it calls for an expansion of corporate power in the hope that it will be used to the benefit of society. If we take the adversarial nature of Hohfeld’s scheme seriously we should be very cautious about taking this bet.

An alternative principle for policymakers or corporate planners wishing to redesign the architecture of corporate purpose in a more stakeholder-oriented manner is instead the following: expand managerial privilege only when the benevolence of the relevant corporate actors is verifiable; otherwise, pair privilege with enforceable duties or countervailing powers.

You can read the current draft of the paper here.