The New Form Agency Problem: Cooperation and Circular Agency

1 Introduction

This paper explores an unsolved issue of the agency problem and adds a new research direction: cooperation between the principal and the agent. Through cooperation between the principal and the agent, the principal has the ability to oppose its own principal, the entity it also acts as an agent for. This situation is not modelled in the traditional agency literature but often occurs in practice. The analysis and inferences of our theoretical model may shed new light on the agency problem, offer ways to improve the efficiency of the participants, and enhance their overall economic benefits.

Traditionally, the agency problem refers to the conflict of interests between the principal and the agent, and is generally solved by rewarding the agent. In addition to rewarding the agent, this research also addresses the agency problem using a novel approach: through cooperation between the principal and the agent. Jensen and Meckling (1976) argued that when the interest pursued by the agent is inconsistent with the interest of the principal because the participants all seek to maximize their own interests, the agent harms the principal. Therefore, the agency problem arises. In the agency problem discussed in this study, in addition to entrusting work to an agent, the principal itself may also act as an agent of another principal. This article refers to this agency problem as the circular agency problem.

There are two major types of principal-agent problems: asymmetric information about types of agents resulting in adverse selection problems and moral hazard problems. The agency problem has been extensively explored by researchers. Although, to the best of our knowledge, the circular agency problem has not been investigated, studies close to our research include those of Grossman and Hart (1983), Maskin and Tirole (1990), and Sappington (1991).

Grossman and Hart (1983) and Mirrlees (1975) found that when the principal maximizes his expected utility and is subject to the first-order conditions of the agent, a solution for the principal-agent problem may not always be obtained, unless the solution to the agent’s maximization problem is unique. This is because when the principal chooses a risk sharing contract or incentive scheme to maximize expected utility, and the expected utility of this agent is at an optimal level, a solution to the principal-agent problem may not always be available. Therefore, Grossman and Hart (1983) developed an optimal incentive scheme that designs the agent’s preference for the income lottery to be action-independent in order to obtain an equilibrium solution.

Maskin and Tirole (1990) first analyzed the agency relationship as a three-stage non-cooperative game: contract proposal, acceptance/rejection, and contract execution. They assumed that the information would not affect the agent’s payoff and obtain a locally unique equilibrium to study different types of principals to motivate the agent’s efforts.

Sappington (1991) explored the frictions that lie at the heart of the incentive problem and studied the optimal response of the principal to these frictions, considering the characteristics of the principal and the agent as they interact with each other. He solved the principal’s task of selecting the best agent considering repeated agency relationships.

While in the above mentioned literature a clear hierarchy between principal and agent is assumed, in many real world situations such hierarchies are not always so clear. With more than three agents involved, sometimes circular hierarchies may occur. A company hierarchy is roughly divided into the board of directors, the manager, the employees, and external shareholders. As an act of corporate governance, the board of directors assigns the manager to invest professional considerations in the company’s operations to handle the company’s business and related matters. The manager then assigns work content to employees, who perform the company’s operations. The professional skills of the manager can be fully used to direct employees to improve company performance, which is then presented to the board of directors. In this agency relationship, the main discussion is about the incomes of the agents. When the manager is not satisfied with the salary given by the board of directors, agency problems will arise because everyone wants to maximize their own interests. In studies of the agency problem, scholars have comprehensively discussed the reward system for agents.

In addition to the above-mentioned agency issues of the board of directors and the manager, other agency issues are worthy of our consideration. Hoping the company does well, the board of directors pays the manager. The external shareholders want to make money by investing in the company. The manager aims to gain a reputation by generating good results for the company, which are observed by the board of directors and communicated to external shareholders. Good performance will make external shareholders more willing to provide capital resources to the company to enable it to expand and grow.

Generally speaking, the manager is an employee who is expected to maximize the interests of the external shareholders. However, a conflict of interest between the manager and the external shareholders occurs because their individual goals are not identical, creating an agency problem. The manager works hard to attract investment from external shareholders. Therefore, the manager is a principal of the external shareholders, and the external shareholders are the agent of the manager.

In commercial transactions, a substantial amount of financial resources, manpower, and goods are continuously invested, resulting in a cycle. The same is true in the operation of a company. The board of directors, the external shareholder, and the manager are both principals and agents in these transactions. They have different perspectives, but they all invest efforts and gain rewards. This study defines the interlocking influence of these layers as “circular agency”.

The purpose of this article is to explore the problems caused by the influence of the roles of the principal and the agent that exist simultaneously in a single actor in the agency relationship. We explore the issue of circular agency by examining the interactions of the board of directors, the external shareholder, and the manager. We examine how these three parties maximize the benefits and resources they are pursuing. We contend that the interlocking agency problems represented by the interactions of these layers are important issues.

Many studies explore the agency problem. Fosu et al. (2016) found that agency problems occur for a variety of reasons, and that when a principal entrusts activities to an agent, the agent at the receiving level of information often knows more about the “efforts” he made to carry out such activities than the principal, creating an information asymmetry. This information asymmetry causes agency problems and affects the company and its value. Chen, Liu, and Zhang (2016) found that under information asymmetry, when reliable information is received, it can alleviate agency problems that may arise, thus promoting the activeness of market transactions. Wang and Wang (2017) argued that under information asymmetry, when one party has private information, it will affect the transaction.

Mu, Wang, and Yang (2017) argued that in the capital structure, the external shareholders and the manager are the main objects of discussion because the agency relationship between the external shareholders and the manager will change due to moral hazards, increasing the cost of the agency relationship. Kuang and Lee (2017) contended that information asymmetry can also increase the probability of potential fraud. Xu, Zhang, and Chen (2018) found that the age of the board of directors and the CEO will have an impact on fraud.

To address the influence of agency problems and thus reduce the possibility of harm, many papers have formulated reward contract mechanisms. Wu, Zhao, and Tang (2014) explored the consideration of the optimal contract in the agency problem with incomplete information. Karim, Lee, and Suh (2018) found that in addition to formulating a reward mechanism, scholars have also proposed various studies that can alleviate agency problems. The CEO can improve company value and shareholder relations through corporate social responsibility (CSR). Jian and Lee (2015) also believed that the CEO’s salary and the CEO’s decision to implement corporate social responsibility are related and will affect each other.

Scholars have also explored the motivation for the objective functions of the three parties. Lee and Oh (2022) studied the impact of mandatory disclosure and management incentives on company investment, finding that if the manager incurs private costs in disclosing information about operations, mandatory disclosure will increase the incentive for the manager to maximize profits. If the disclosure costs are higher and the manager has less ownership, shareholders will have more incentives to allow the manager to take care of his private interests. Therefore, an agency problem between the manager and the shareholders arises. Oh and Park (2016) studied the impact of market competition on corporate governance. They found that when the market is more competitive and the manager’s ownership ratio is low, and his control over the company is high. Therefore, there is an agency problem between shareholders and the manager. Oh and Shin (2020) studied the level of market competition under which shareholders allow weaker management of companies in response to market competitive pressures. They found that when the threat of competition was greater, shareholders with higher institutional ownership and boards took more aggressive decisions on compensation committees, that is, their aggressive behaviors represented the strategic decisions of shareholders. Furthermore, under high competitive pressure, they found a positive correlation between the adoption of entrenchment provisions and the future performance of the company. Therefore, an agency problem between the external shareholder, the board of directors, and the manager would arise.

The issue of collusion in the circular agency is explored in the following sections. Subsection 2.1 considers the cooperation between the shareholder and the manager, Subsection 2.2 considers the cooperation between the manager and the board of directors, and Subsection 2.3 considers the cooperation between the board of directors and the shareholder.

This study considers the collusion of two participants. This collusion group has an agency problem with the third participant. At first glance, it appears that the interaction of a “circular agency” is not taken into account. It may appear that these three subsections should be treated as two-party agency problems rather than a collective “circular agency” problem. However, these are “circular agency” issues since they relate to discussions of simultaneous power transitions between colluding groups of stakeholders.

In Subsection 2.1, the cooperation between the manager and the shareholder is discussed. The manager of this collusion group is the principal, that is, the manager of this collusion group is the leader, while for the board of directors, the manager is the agent, meaning that the board of directors becomes the principal of the colluding group.

In Subsection 2.2, the cooperation between the board of directors and the manager is discussed. The board of directors of this collusion group is the principal, since it is the leader. For the shareholder, the board of directors is the agent, making the shareholder the principal of the colluding group.

In Subsection 2.3, the cooperation between the shareholder and the board of directors is discussed. As the leader, the shareholder in this collusion group represent the principal, whereas, for the manager, the shareholder is the agent, making the manager the principal of this colluding group.

To address this complex, three-cornered agency problem in which actors are simultaneously agents and principals, we have coined the concept of “circular agency”. Our model of the interactions of this circular agency is described below.

2 Basic Model

The circular agency design model for the three participants is as follows. A circular agency consists of three agents: The (representative) shareholder (principal 1 and agent 3, simultaneously), the board of directors (agent 1 and principal 2 simultaneously), and the manager (agent 2 and principal 3 simultaneously).

The (representative) shareholder is a principal of the board of directors because the objective of the shareholder (principal 1) is to obtain capital gains and invest funds in the board of directors (agent 1). The board of directors is a principal to the manager because the objective of the board of directors (principal 2) is for the company to have good financial performance, and it pays salaries to the manager (agent 2). The manager is a principal to the shareholder because the objective of the manager (principal 3) is to be able to develop her talent in a large-scale company and to build a reputation (this reputation is due to the investment of funds by the shareholder (agent 3) to expand the scale of the company).

In addition, we assume the manager is not also a member of the board of directors. Our model does not consider independent or external directors. At the same time, the manager is a representative manager, that is, the representative manager of all managers. In the same way, the external shareholder is a representative shareholder, that is, the representative shareholder of all shareholders.

This study defines the compensation of principal 1 (the shareholder) as π ¹, the compensation of principal 2 (the board of directors) as π ², and the compensation of principal 3 (the manager) as π ³. We assume that Principal 1 (the shareholder) wants to obtain capital gains through investment, and invests funds in the company under the guidance of the board of directors. We then assume that Principal 2 (the board of directors) wants the company to show good financial performance to the external shareholder, demonstrating that the company is worth investing in. The financial performance here refers to a book surplus but not necessarily a true surplus. Therefore, the salary is given to the manager who is entrusted with the management of the company. We then assume that Principal 3 (the manager) wants to develop her talent in a large-scale company and wants to build a reputation. This reputation is generated by the shareholder, who can expand the scale of the company through large capital investments. The manager contributes her professional knowledge and strives to manage the company.

The payoff of principal 1 (the shareholder) is π ¹. In this payoff, the shareholder pays funds to invest in company x to obtain the capital gain distribution b. The manager seeks to gain reputation c. Therefore, the manager works hard, causing disutility v e , and uses her professional skills to allow the company to obtain financial performance g m for presentation to the shareholder, to attract him to invest capital x to support the management of the company’s operations, and build a good reputation c, where m is real surplus. Thus, the real earnings of the manager’s efforts are practice the actual net income of the company, where v e above is an addition to the payoff of the shareholder. When the effort of the manager is in the interests of the shareholders, the more the manager can evade moral hazards while being the manager, the more the manager can avoid the moral hazard of the manager. If the board of directors emptied the company’s assets due to the manager’s mismanagement, the manager’s effort is an addition to the shareholder’s, where v e is assumed to be equal to e ²/2. Assuming that the effort invested by the manager is the same as the agent’s income, and e ∈ e | 0 ≤ e < ∞ , the reason the manager’s reputation c is a deduction for the shareholder is that the reward cost c paid by the shareholder to the manager’s reputation c is equal. Since the effort of the manager e is fully reflected in the agency income, the agency income is equal to e.

The payoff of principal 2 (the board of directors) is π ². In this payoff, the board of directors receives the invested capital x from the shareholder and assigns the manager to use it to obtain the financial performance g m , where g m is the book surplus, but not necessarily the real surplus. The board of directors gives the manager a salary s m . Let s m = β + α ⋅ m , where β is the basic salary, and α is the performance salary rate. With good performance, the board of directors can also pay dividends (or capital gains) to the shareholder b. Although the shareholder will invest funds x in the company, the shareholder is not on the board of directors. However, the equity x invested by the shareholder is not the profit of the board of directors, though the payoff of the board to directors will increase due to this investment.

The payoff of principal 3 (the manager) is π ³. In this payoff, the manager expects to develop her talent in a large firm and gain reputation c. With her hard work v e , she can obtain salary s(m). The manager also works hard to make the financial performance g m look good. This cost of the manager is assumed to be the same as the value of g m in the financial performance. An objective of the manager is that the shareholder can have confidence in the company and invest more funds x, giving the manager more resources to expand the size of the company, building her reputation.

In real-world examples of the three-party agency problem in the company, the personnel hierarchy in public or OTC companies is diverse. Therefore, the membership of the company will vary with the organization of the company. However, the main decision makers of the company can generally be abstracted into three participants: the shareholder, the board of directors, and the manager. For the sake of simplicity, we ignore other personnel in the company. The nine variables in this study are defined below. First, the capital gain paid by the board of directors to the shareholder is b, which in practice is the stock dividend obtained by the shareholder investing in the company. Second, the manager’s performance salary rate is α, which in practice is the manager’s share of the company’s earnings. Third, the basic salary is β, which in practice is the fixed salary the manager receives from the company. Fourth, the manager’s salary is s(m), which in practice is the total salary that the manager receives from the company, where s m = β + α ⋅ m . Fifth, the capital invested by the shareholder is x, which in practice is the total investment of the shareholder in the company. Sixth, the reputation of the manager given by the shareholder is c. In practice, this is the degree of satisfaction of the shareholder with the performance of the manager. Seventh, the effort of the manager is e. In practice, this is the ability or effort of the manager in running the company. Eighth, the real surplus shown by the manager is m, which in practice is the actual net income of the company. Ninth, the book surplus is g m , which in practice is the value of the company’s net book earnings.

3 Conclusions

This paper addresses the lack of discussion of cooperation issues in the traditional agency problem. After the principal cooperates with its agent, a new cooperative formation can compete with the principal of the cooperating principal. This paper solves the equilibrium solutions for this agency problem. Thus, this paper expands on the original theory of agency to formulate a new construction of the agency problem.

In the case of the three principals of the circular agency who are also agents, the propositions discuss the agency problem between the manager and the shareholder, the board of directors and the manager, and the shareholder and the board of directors. When rights and responsibilities are combined, cooperation is very likely to occur to oppose the principal to gain greater self-benefits. Therefore, the purpose of this study is to explore the equilibrium between the cooperation of the parties in the case of circular agency.

Proposition 1, Proposition 2, and Proposition 3 depict the cooperation problem of the circular agency. In Proposition 1, the principal is the board of directors, and the manager and shareholder cooperate with each other as the agent. In Proposition 2, the principal is the shareholder, and the agent is the board of directors colluding with the manager. In Proposition 3, the principal is the manager and the agent is the shareholder who cooperates with the board of directors.

When the board of directors (principal 2) is the principal, the manager and the shareholder cooperate and act as an agent. When the shareholder (principal 1) is the principal, the board and the manager cooperate and act as an agent. When the manager (principal 3) is the principal, the shareholder and the board of directors cooperate and act as an agent. In the agency problem, each participant maximizes their own interests, prompting the agents to cooperate. This study uses a theoretical model to find that in the case of cooperation, when the board of directors is the principal and the manager and the shareholder in cooperation, the shareholder will receive dividends, while when the shareholder is the principal, the board of directors and the manager are in cooperation, and the manager will also gain the reputation given by the shareholder.

The theoretical model of circular agency in this study represents a pioneering exploration of a novel perspective on agency theory using a theoretical model and analysis. It is hoped that this research can open up new directions in agency theory.

This paper explores the unsolved issues of the agency problem and adds a new research direction, circular agency theory, and analyzes the cooperation between the principal and the agent. This situation is not modelled in the traditional agency literature, but it often occurs in practice. The limitation of this study is that since the number of participants in different companies will differ across their varied hierarchical structures, the equilibrium results obtained will also be different. The model of this study is still somewhat restrictive, as it assumes that there are only three participants. Future researchers can explore other generalizations and extensions across different company types. Furthermore, we also suggest that future research directions can be achieved through the cooperation of a single participant with its principal and its agent, and through the negotiation of these two cooperative groups to further form the equilibrium.