by Professor Adam Meirowitz, David Eccles School of Business, University of Utah, Dr Shaoting Pi, Research Associate, Cambridge Centre for Finance and Cambridge Endowment for Research in Finance, University of Cambridge, and Dr Matthew C. Ringgenberg, David Eccles School of Business, University of Utah
Voting plays an important role in corporate governance. Shareholders vote to elect members of the board of directors, and they vote on proposals that may directly affect the actions of the firm. Similarly, members of the board of directors vote to appoint the chief executive officer and vote on a variety of firm policies. Traditionally, these stakeholders broadly agreed on the objective of the firm – maximise firm value – and this one-dimensional objective simplified voting. Yet more recently, a number of academics, practitioners, and regulators have argued that firms ought to care about more than just value. While concern over environmental, social, and corporate governance (ESG) is exploding, there has been little conceptual work evaluating whether this shift in the scope of what matters to shareholders and board members impacts the performance of firm governance.
How does one conceive of firm governance in the presence of ESG concerns? One defensible position is that now board members and shareholders have a multi-dimensional objective function; they must make tradeoffs between value creation and socially responsible behaviour. (eg, B’enabou and Tirole (2010), Hart and Zingales (2017)). This departure itself can be consequential. With agreement on just maximising firm value, we might have thought of governance challenges as emerging from just differences of opinion about the best way to enhance firm value. Now shareholders or board members may genuinely disagree about how much value they are willing to give up in order to advance a social objective (and they may disagree about the best way to achieve any particular balancing of these goals). But, on further consideration, we find that this perspective itself is insufficiently nuanced. As the name, Environment, Social, and Governance, suggests, ESG may be multidimensional. In other words, we may think of board members facing disagreements over both the margin between value creation and socially responsible behaviour as well as a potentially large number of margins between different aspects of socially responsible behaviour. This is the perspective that we take. In this paper, we build a theory of corporate voting over policies that impact value as well as ESG dimensions. Our analysis fleshes out how the movement from concern over value to concern over value plus ESG impacts. In particular, we find that a narrow or focused notion of ESG can result in minimal challenges, while a broader or multi-dimensional notion may degrade governance in important ways.
Building on the literature on social choice theory, we theoretically examine voting for corporate policies when voters face a trade-off between maximising firm value and one or more social policies (for example, reducing pollution and increasing employee satisfaction). Arrow (1951) famously shows that no method of aggregating preferences will satisfy a small set of naturally satisfying axioms. However, subsequent literature shows that stable choices can emerge under various restrictions on voter preferences. A well-studied restriction is the case of single-peaked preferences which is satisfied if the feasible policies can be arranged in a single dimension and on this dimension, all agents’ preferences are quasi-concave (informally, monotone or tent-shaped). When this restriction is satisfied, it is possible to identify choices that seem to reflect the will of a majority or aggregate preferences. We show that a number of challenges arise when the firm objective function is expanded to incorporate social policies. Interestingly, we show that adding just one social dimension does not lead to additional problems. After accounting for a feasibility constraint, preferences over firm value and one social dimension still satisfy an order restriction, and social choice is well-behaved. However, when more than one social dimension is present, challenges emerge – we show that the resulting choices are likely to be sensitive to institutional features that may vary or be difficult for investors to understand and track. In other words, our findings show that decision-makers may be able to add one social dimension to a firm’s objective function; however, the quality of the firm’s governance will decline if decision-makers care about too many objectives.
A simple example helps to illustrate our key findings. Imagine a firm with three possible policy choices. The firm can implement a policy, P, that maximises firm value or a policy G that is less profitable but environmentally sustainable or a policy E that is less profitable still but mandates ethical treatment of workers. Consider three investors (or three board members) who are charged with making the policy choice, denoted as investors 1, 2, and 3. Suppose that investor 1 cares only about firm value and thus orders the alternatives by expected firm value: P, G, then E. Suppose that investor 2 most prefers to protect the environment, but would still rather support the ethical treatment of workers over just maximising firm value and so ranks the alternatives, G, E, then P. Finally, suppose investor 3 cares about the ethical treatment of workers but is not willing to sacrifice returns for environmental policies and thus ranks the alternatives E, P, and then G. If any two of these policies are offered, a stable choice will emerge. In particular: given a choice between E and P, E wins. Given a choice between P and G, P wins. And given a choice between G and E, E wins. However, if all three policies are offered, none of the alternatives beats the other two alternatives. While E beats P, G beats E, yet P beats G. As a result, when the three policies are offered, none of them is naturally preferred or stable under majority rule. The ultimate policy choice may thus depend on additional and less obvious features of the institution. This creates additional challenges for an investor who might face serious uncertainty about the firm’s likely policy choice.
What are the implications of this finding? If the firm faces more than two dimensions in its objective function, there is generally no natural policy choice. As a consequence, the choices that emerge will depend on the process by which policies are proposed, and the volatility of firm choices will tend to increase. This simple example illustrates a deep and practical concern about preference aggregation. Although majority rule (and other stronger super-majority rules) is not immune from Arrow’s impossibility theorem, there are still compelling reasons to use them. In particular, when preferences are single-peaked (or more general satisfy order restriction), the majority rule is known to be well-behaved; many systems that involve fairly decentralised proposal rights and majority voting will tend to select policies that are quite responsive to the preferences of the so-called median voter. But when preferences do not satisfy these types of restrictions, the outcomes can depend heavily on seemingly subtle institutional features. Whether a policy-making domain exhibits enough preference diversity for this problem to become important is an applied question, that to date, has not been extensively studied in finance contexts (such as choosing socially responsible corporate policies). Our paper fills this void. We show that when a group of voters have monotone preferences over at most two dimensions (firm value and one social dimension) and face a natural feasibility constraint, then we may think of the preference aggregation problem as nice or well-behaved. But this no longer holds with three or more dimensions.
 DeMarzo (1993) shows an exception: if markets are incomplete, then investors may also disagree on how to maximise firm value, which complicates the public choice problem.
 A large literature examines the challenges of corporate governance when the goal is to maximise firm value. For example, there is extensive work on the agency conflict that arises between investors and managers when ownership and management are separate. See, for example, Berle and Means (1932); Jensen and Meckling (1976); Demsetz (1983); Admati, Pfleiderer, and Zechner (1994); Burkart, Gromb, and Panunzi (1997); Maug (1998).
Admati, A.R., Pfleiderer, P. and Zechner, J. (1994) “Large shareholder activism, risk sharing, and financial market equilibrium.” Journal of Political Economy, 102(6): 1097–1130
Arrow, K. (1951) “A difficulty in the concept of social welfare.” Journal of Political Economy, 58: 328–346
B’enabou, R. and Tirole, J. (2010) “Individual and corporate social responsibility.” Economica, 77: 1-19
Berle, A.A. and Means, G.C. (1932) The modern corporation and private property. New York: Macmillan.
Burkart, M., Gromb, D. and Panunzi, F. (1997) “Large shareholders, monitoring, and the value of the firm.” Quarterly Journal of Economics, 112(3): 693–728
DeMarzo, P.M. (1993) “Majority voting and corporate control: the rule of the dominant shareholder.” Review of Economic Studies, 60(3): 713–734
Demsetz, H. (1983) “The structure of ownership and the theory of the firm.” Journal of Law and Economics, 26(2): 375–390
Hart, O. and Zingales, L. (2017) “Companies should maximize shareholder welfare not market value.” Journal of Law, Finance, and Accounting, 2: 247–274
Jensen, M. and Meckling, W. (1976) “Theory of the firm: managerial behavior, agency costs and ownership structure.” Journal of Financial Economics, 3(4): 305-360
Maug, E. (1998) “Large shareholders as monitors: is there a trade-off between liquidity and control?” Journal of Finance, 53(1): 65–98
Meirowitz, A., Pi, S. and Ringgenberg, M.C. (2022) “Voting for socially responsible corporate policies.” Social Science Research Network (SSRN) paper.