Print
E-MailLorenzo Garlappi is Associate Professor of Finance, Ron Giammarino is Professor of Finance, and Ali Lazrak is Associate Professor of Finance at the University of British Columbia Sauder School of Business. This post is based on a recent article, recently published in the Journal of Financial Economics, by Professor Garlappi, Professor Giammarino, and Professor Lazrak.
The word “corporation,” derived from the Latin corpus, or body, refers to “a body formed and authorized by law to act as a single person.” [1] The study of corporate decisions typically models the corporate body as either (i) a single person, e.g., a manager, who maximizes expected utility with respect to a unique prior belief, or (ii) a decision-making group consisting of utility-maximizing individuals with identical prior beliefs, albeit possibly differentially informed. In reality, corporate boards and management teams are examples of groups where individuals with different opinions must collectively decide, as a single legal person, what the corporation is to do, often in the face of dramatically different views about whose model of the world is correct.
In this article we study corporate decisions made by a group for which the “common prior” assumption does not hold. If group members must collectively make a decision on behalf of the corporation but have heterogeneous priors, the group is de facto a “multi-prior” decision maker, even if each individual member has a single-prior. The group, therefore, faces an “ambiguous” decision problem (e.g., Ellsberg, 1961). Although the term “ambiguity” refers typically to a single decision maker who holds multiple priors over outcomes, it is suggestive of the environment faced by a group with heterogeneous beliefs whose members disagree on the probability of the states of the world.
In our analysis, all group members have preferences described by Savage’s (1954) Subjective Expected Utility model with identical utility functions. Consumption is common across group members who differ only in their subjective beliefs about the likelihood of future outcomes. We assume that, when faced with choices that are not unanimously ranked by all its members, the group invokes a utilitarian governance (or aggregation) mechanism.
According to this mechanism, group decisions are obtained through a fictitious expected utility agent whose beliefs are the linear combinations of individual beliefs, with weights that are constant over time. The utilitarian mechanism can be thought of as a reduced-form version of the complex interplay of legal, political, and economic forces that characterize the corporate governance process. Viewed under this light, the weight attached to the utility of a group member in the utilitarian mechanism can be interpreted as the influence that the individual has on corporate decisions, be it through personal attributes, social status, or legal power.
We study the choices of a group that has to decide whether or not to invest in a new project and, upon investment, whether to continue or abandon after receiving a signal about the project’s likelihood of success. Our analysis delivers three main theoretical implications for dynamic decision-making, corporate investment, and security design. First, we show that the utilitarian mechanism is dynamically inconsistent, that is, the ex ante ranking of two alternatives can be reversed after the group members learn about payoff irrelevant states of the world. Intuitively, dynamic inconsistency arises because learning may induce shifts in the relative influence of a group member on the group decision. As a result, members of a group who are relatively uninfluential in a decision before learning may become more influential after learning.
Second, we show that dynamic inconsistency can lead to a novel form of investment inefficiency where all members of the group, despite their different views of the world, agree that investment is best but nevertheless collectively decide not to invest. Intuitively, some group members who would support one future operating choice over another recognize potential conflicts that may arise as a consequence of future learning. Rationally anticipating how the conflict will be resolved in the future, the group members that disagree with the expected resolution end up opposing the initial investment.
Third, we show that allowing for trading typically mitigates the inefficient underinvestment that occurs in the absence of markets although, in some cases, it leads to inefficient overinvestment. Specifically, when group members can trade among themselves, the member with the highest valuation will take over the firm and invest. Moreover, when group members collectively trade with outside investors, they can resolve the underinvestment problem by issuing a security that can change the firm’s payout across different states of the world in a way that brings unanimity in operating decisions. Intuitively, the expected future conflict that causes the initial inefficiency comes about when a future decision becomes relatively more important to some group members after learning. We show that contracts can be designed to eliminate future disagreement. This finding highlights an entirely new role for financial contracting in neutralizing conflicts that may arise among group members with heterogeneous beliefs.
We contribute to several strands of the finance literature. First, we explicitly recognize the ambiguity-like nature of corporate decisions undertaken by a group of individuals with heterogeneous beliefs. Second, we highlight a novel implication of belief heterogeneity on economic decisions. While the existing literature shows that differences in beliefs in a market-mediated environment may lead to speculative trading (see e.g. Harrison and Kreps (1978) and Scheinkman and Xiong (2003)) and overinvestment (see Shleifer and Vishny (1990)) our findings suggest opposite results: when decisions have to be made collectively and trading among group members is not allowed, differences in beliefs may lead to underinvestment.
Third, our work emphasizes disagreement as a new source of inefficiency. Prior work in corporate finance shows how inefficiency may result as a consequence of information asymmetry and/or conflicts of interest between existing debt and equity holders (e.g., Myers, 1977), or between old and new equity holders (e.g., Myers and Majluf, 1984). Our finding shows that differences in beliefs lead to inefficient underinvestment when decisions are collectively made on behalf of a corporation even when all agents receive the same signals, the interests of all agents are aligned, and all agents have a common claim to the payoff of the investment. We further show how contracting can help mitigate such inefficiencies.
Our results provide a new framework for studying diversity within an organization. Empirically, several studies have looked at the relationship between corporate decisions and the heterogeneity of decision makers in terms of gender, age, education, etc. It is often conjectured that observed behavior can be explained in terms of differences in risk aversion, overconfidence, or culture. Our results suggest a new explanation for these empirical results based on the aggregation of heterogeneous beliefs within the corporation.
While our analysis is robust to various enrichments of the proposed model, we recognize that it rests on three critical assumptions: (1) collective decision making, (2) heterogeneous beliefs, and (3) the absence of trading. Despite their limitations, we believe that these assumptions accurately capture the environment of many corporate decisions.
The complete article is available here.
Endnotes
1Merriam Webster Dictionary.(go back)
