agency theory, financial

Introduction

      in organizational economics, a means of assessing the work being done for a principal (i.e., an employer) by an agent (i.e., an employee). While consistent with the concept of agency traditionally advanced by legal scholars and attorneys, the economic variants of agency theory emphasize the costs and benefits of the principal-agent relationship. While a beneficial agency cost is one that increases a shareholder's value, an unwanted agency cost occurs when management actions conflict with shareholder interests. Such would be the case when managers put their own interests ahead of an owner's interests (e.g., manipulating short-term earnings at the expense of long-term performance in order to receive a bonus). Ongoing analyses of agency costs are a common managerial tool, especially in corporations that are managed by nonowners, because they serve to indicate whether—or how well—a manager (agent) is fulfilling his fiduciary obligation to an owner (principal).

Theoretical development

      Contemporary applications of agency theory were advanced with the publication of Theory of the Firm: Managerial Behavior, Agency Costs, and Ownership Structure (1976), published in the Journal of Financial Economics by financial economist Michael C. Jensen and management theorist William H. Meckling. Building on earlier work by the American economists Ronald Coase (Coase, Ronald), Armen Alchian (Alchian, Armen A.), and Harold Demsetz, Jensen and Meckling developed an economic model specifically designed to capture the essence of the principal-agent relationship.

      Consistent with the legal understanding of agency, Jensen and Meckling described the agency relationship as a contract (explicit or implied) in which one person, the principal, hires a second person, the agent, to perform some action. In such cases the principal formally delegates decision-making authority to the chosen agent.

      Jensen and Meckling began by assuming that each party to the contract consistently chooses those actions that are most likely to maximize his own expected utility (in other words, both agent and principal always act so as to promote their own self-interest). Although an agent's motivations may include the desire to work hard to achieve the principal's goals, he may also be motivated by a desire to maintain the prestige or perquisites associated with the job, such as well-appointed offices and the use of corporate jets (all of which can be viewed as an economic loss from the principal's perspective). Although the assumption that both parties seek to promote their own self-interest is controversial among economists, a fact that Jensen and Meckling acknowledge, it remains the central tenet of agency theory.

      Jensen and Meckling emphasized the precise nature of the costs inherent in all agency situations by isolating three components: the costs incurred by the principal to monitor the agent's behaviour; the costs (such as bonding expenses) incurred by the agent to guarantee the quality of his actions; and the cash value of any loss in utility experienced by the principal that results from the agent's self-interested behaviour. The last component, known as a “residual loss,” occurs whenever the actions that would promote the self-interest of the principal differ from those that would promote the self-interest of the agent, despite monitoring and bonding activities. Depending on the situation, the costs of agency can be quite significant in relation to the size of the project.

      Viewed from a perspective of rational choice (political science) (i.e., that individuals make decisions and take actions that bring advantage to them), the principal can expect the agent to behave in a self-interested way. In other words, there is a high likelihood that the agent will place greater priority on actions that will serve the interests of the agent rather than the principal. The principal can therefore take strategic steps to limit the damage caused by the agent's self-interested behaviour. Common approaches include defining job expectations and writing contracts in a way that encourages the desired behaviour while limiting any divergent (e.g., costly) behaviours; these frequently involve performance incentives, including rewards for good performance and penalties for poor performance. Next, the principal may wish to hire a third party to monitor—or, at minimum, obtain a sampling of—the agent's actions. Retail companies, for example, can hire mystery shoppers to test the performance of sales personnel. As long as the cost of hiring the monitor remains lower than the additional benefit the principal gains by scrutinizing his agent's behaviour, the principal will find it in his interest to hire a monitor. Correspondingly, agents may discover that bonding costs can be well worth the expense if they increase the agent's value to the principal, because bonding provides a recognized means of ensuring behaviour consistent with the principal's desires. Agents who bond themselves can often obtain higher levels of compensation compared to agents who do not incur bonding costs. (See also guaranty and suretyship.)

Applications

      Agency theory entails a number of specific and testable empirical hypotheses. For example, it has been used to explain why stockholders are willing to accept managerial behaviour that does not maximize the value of the firm. It provides insight into the reasons why managers voluntarily produce audited financial reports on an annual basis. In addition, the agency perspective can explain why ownership structures differ across industries (such as steel and software), and it can cast light on the restrictions (such as those found in bond covenants) imposed by creditors on managerial actions.

      In the late 20th century, economists and management scholars broadened the basic agency model to include questions about externalities (negative effects stemming from the company's course of business, such as air pollution), the ethical assumptions underlying the agency model, and the social costs associated with private dishonesty. In such cases the question of agency was raised from the level of individual agency to the level of the organization itself, whereby the organization's self-interested behaviour is weighed against the collective social impact of that behaviour. Organizational behaviour in these cases is explained as a function of (1) the divergent interests of the principal (i.e., society) and agent (i.e., the organization) and (2) the limitations on the principal's ability to fully observe agents' actions.

      Although agency theory has been used chiefly to analyze the behaviour of economic actors in the for-profit business sector, many of the insights of this perspective can be applied to any situation in which cooperative behaviour is necessary to achieve a desired goal. Consequently, the theory can be used to understand economic choices made by nonprofit organizations such as universities (university), philanthropic foundations (philanthropic foundation), and trade unions (trade union).

Moses L. Pava

Additional Reading

Descriptions of the agency model are provided in Michael C. Jensen and William H. Meckling, “Theory of the Firm: Managerial Behavior, Agency Costs, and Ownership Structure,” Journal of Financial Economics, 3:305–360 (1976). A more analytical perspective can be found in Bengt Holmstrom, “Moral Hazard and Observability,” The Bell Journal of Economics, 10:74–91 (1979). Oyvind Bohren, “The Agent's Ethics in the Principal-Agent Model,” Journal of Business Ethics, 17:745–755 (1998), discusses the agency model from an ethical perspective.Moses L. Pava

* * *