Received 6 June 2016; accepted 16 August 2016; published 19 August 2016
Firstly, under the assumption of neoclassical economics, since the market price mechanism, economic agents (manufacturers, consumers and government) have complete abilities of information receiving and processing to support their choices and behavior, so that the economic behavior is completely certain. Secondly, in the view of technology, “black box” theory set firms as completely effective black boxes: everything runs perfectly inside, the information is sufficient, and entrepreneurs always want to maximize firms’ profits. However, there is a huge gap between these assumptions and the reality. In this situation, expected utility theory which is established by Von Neumann states that the decision makers choose between risky or uncertain prospects by comparing their expected utility values, i.e., the weighted sums obtained by adding the utility value of outcomes multiplied by their respective probabilities. An Morgenstern in 1944 and Choice Theory that is set by Gerard Debreu in 1959 and the central aspect of Choice Theory is the belief that we are internally, not externally motivated. While other theories suggest that outside events “cause” us to behave in certain predictable ways, Choice Theory teaches that outside events never “make” us to do anything, which is utmost of importance. On the basis of this, information economy and agency theory appeared. According to the definition from Jeason and Mecling (1976), principal-agent means a contract which is explicit or implicit, put it in other way, one or more actors hire the other actors to provide service; at the same time, the front grants the latter some political power. The amount and quality of the service decide how much money the latter can get. Above all, the authorized actor is principal and the grantee is agent.
The author of “The Wealth of Nations”, Adam Smith (1776) proposed that when business managers are not business owners, carelessness and waste is inevitable. What he discussed is agency problem between shareholders and managers when ownership separates from controlling. Berle and Means (1932) pointed that, under the circumstance of ownership separation, there are inconsistent benefit between company managers who do not have stake and fixed shareholders. And this situation will lead that this kind of managers holds corporate controlling power firmly, so managers are likely to use corporate resources for personal gain rather than the maximization of corporate value. Since the separation of ownership and control of enterprises in the economic activity is ubiquitous, so it has been a widely held point of departure of Enterprise agent problem. Jeasen and Meckling (1976) defined agency relationship as a contract and it raised real concern about the operator agency. In this contract, the principal grants agent to fulfill certain services, including entrusted decision-making powers to the agent. Because information asymmetry and inconsistency of both aims, there is a conflict between principal and agent, which is the agency problem, the loss of the enterprise value caused by the agency problem is agency costs, including loss of surplus value, monitoring costs and guarantee costs.
The core of principal-agent theory is to solve the problem of restrictions on the behavior of the agent under condition of asymmetric information and conflict of interest. He (2009)  proposed that with reducing costs, the behavior of the agent can provide a more performance for the commission people grab more benefits. And the problems to be solved here are as follow: 1) reducing unnecessary agency costs, 2) improving the performance of the organization. This two aspects lie in forming the agent’s behavior. Ma (2014)  discovered in recent years, the core of agency theory puts more concern on how to design to an effective incentive and restraint mechanisms to control agent’s behavior to minimize the cost and maximize their own benefits at the same time of resolving the conflict interests and asymmetric information.
2. Elaboration of Management Mechanism
Different scholars have their own point of view about how to reduce the cost of agency. To sum up, there are the following four main ideas. Firstly, reducing the cost from adjusting the managers’ payment structure. Designing effective incentive program so that the agent has a residual claim will help improve operators themselves and members of the business the “comprehensive initiative”, making the goal of principal and agent as consistent as possible. Secondly, to improve the corporate governance structure to reduce agency costs, namely the creation of decentralized type of organization, the establishment of risk-sharing incentives. Thirdly, the manager’s job, strict supervision and accurate evaluation. Fourth, the use of market competition mechanism gives the agent more pressure and constraints on their behavior. Competition in product markets prompts managers to save money. Above all, the fact is to encourage and overview the agent to guarantee they can work hard to realize the expected goals, so that principal and agent can get win-win in game.
Jensen and Meekling (1980) considered that viewing the firm as a network which is combined with contract relationship can do some help to understand that business is not a man but the product of law, it is a complex central process, in this process various conflicting objectives reached a balance in the framework of a contractual relationship. This is a logical thinking that can be used to develop research about contract relationship in business. According to this logical thinking, a series of effective management mechanism come out. They can be divided to two categories: integrated excitation method and separated monitoring method.
2.1. Integrated Excitation Method
Principal and agent always want to maximize their own benefits. But the benefit of principal depends on the cost of agent, depends on how hard work the agent does. While the benefit of agent is equal to the cost of principal, Liu and Jiang (2006)  discussed it is the conflict filed that exists between two actors’ relationship . Therefore, most of researches propose that there are two ways to build the connection between two actors. One is to set indirect relationship through weighing performance payment. The other one is to build direct relationship through dividing company shares. Finally the goal of integrated excitation can be realized.
2.1.1. Enterprise Performance Pay Makes the Agent of Interest and Shareholders Linked
From the point of view of both sides, the relationship between the principal and agent is a kind of contract relationship which is entrusting and realizing at the same time, two actors must react to each other to maintain this relationship. From the perspective of one side, the principal entrust parts of decision power to the agent in pursuit of performance, the agent shoulders related responsibilities in pursuit of his own payment In this contractual relationship, the first step for the two actors to meet the needs is to meet each other’s needs, so that the agent must work hard to improve business performance, to satisfy principal’s demands, meanwhile the principal must give corresponding pay. Put it in simple, on a contractual basis there is a “supply and demand” relationship between the two actors (Figure 1).
In the pursuit of their own interests, the best situation is that two actors can meet their needs. In practice, however, their interests cannot be completely satisfied in same level. But at the same time, the condition of sacrificing the interests of the other actor is a great temptation for any actor to get more revenue. Many different domestic and foreign researchers proposed that the relationship between agent and principal is in positive correlation and it can be coordinated.
In United States and Japan, the level of CEOs’ payment and company performance are positive correlated. Chakra borty and Sheikh (2009) set turnover as an indicator of corporate performance, and found that high levels of the agent incentive compensation can drive firm to achieve high corporate performance.
Chen (2010)  started to explore how the mechanism affect agency costs with a variety of family businesses, further obtained professional managers, the agent, their pay mechanism can put more significant effect on reducing agency costs and improving business performance. Luo and Chen (2010)  used accounting surplus comparative study of incentive pay to conduct a study and the results showed a positive correlation between the two actors, and if different levels of the agent appear better regulatory constraints mechanism, the agents’ opportunistic behavior can be reduced greatly. Jiang (2014)  conducted an empirical research on how listed companies executives’ pay affects corporate performance in short and long term, and they concluded that the performance of listed companies and executives’ payment are in positive correlation, but specifically, the impact of monetary remuneration is more significant in the short term, long-term impact is not so good. On the basis of analyzing of real estate in Shanghai and Shenzhen A-share listed companies within a period of three years (2011-2013) sample data, Yao (2015) indicated that listed real estate investment company executive pay and investment benefits are in significant positive correlation.
Different scholars have come to different conclusions about researching the relationship between the interests of the principal and the agent remuneration: there is not perfect positive correlation between them.
Fang (2010)  proposed that there exists adhesive relationship between agent payment and performance. That is, their salary will increase with performance showing growth, but will not make performance decline. Zuo and Tang (2010)  pointed out that agent incentive pay has hysteresis effect, but also concluded that the relationship between executive pay and corporate performance is in a positive correlation, and the relationship can be described with a U-shaped line which is high right low left. That is, at the initial stage of the work, high pay
Figure 1. Performance pay makes the agent of interest and shareholders linked.
can encourage the agent to work hard, but lately, too high salary can inhibit agents.
“Fat cat” phenomenon which is proposed by Jesse and Lucian (2005)  reflects that, to some degree, payment cannot reduce the agents speculation behavior and agency costs, but it is an effective means to eroding the shareholders’ benefits and emptying Listed Companies. Yang and Wang (2010)  pointed out in the long term, the current domestic compensation s, particularly the payment in the form of currency will lead that the agent behavior in short-term exploits the long-term interests of the enterprise.
2.1.2. Connect the Principal’s Interests with Shareholders by Ownership
Adjusting or changing the agent’s compensation structure is an effective way to reduce the cost of agent. Stock and option themselves have a risk. Their essence is to allow business operators have some residual claims and shoulder the corresponding risks. They endow the stock owner to take sharing of company’s profits and the remaining is requested by authority in proportion to the number of shares held by stock owner. Integrating the principal and the agent is different from the above remuneration “targeting”. It can help to improve the relationship between the principal and the agent (Figure 2).
In the “integration”, the impact of the agent stock holdings on the overall benefits can be divided into “convergence of interests” and “entrenchment” effects. These two effects are both ends of the “U-type” curves, and entrenchment effect is “U-type” left: agent motivation and interests of the client show the opposite effect; convergence of interest effect is “U-type” right: agent motivation and interests of the client show the same role.
Fama and Jensen (1983)  believe that the inconsistency in goal between the principal and the agent, asymmetric information, incomplete contract and liability risks, could lead agent to pay more attention to their own interests at the expense of the interests of shareholders, resulting in a conflict of interest between the two. Giving incentive stock options to the agent can tie both interests up, showing a “convergence of interests” effect. Jensen and Meckling (1990) analyzed that increasing agent stake can lead to increase value of the company. That is, when equity relative dispersion and decreasing the percentage of agent ownership will lead to the weakening of its innovative power, leading to a further decline of value of the company. Through researching the relationship between the 100 major UK executives of listed companies financial incentives and corporate performance, Conyon and Sadler (2001)  pointed out that the executives’ effective ownership changes with corporate performance in same direction. Based on the research of 2006-2008 years in the SME board listed 102 companies, Li and Zhao (2010)  found that a measure of managerial ownership and corporate performance such as return on equity (ROE), each EPS (EPS) and net assets per share (NAPS) in different degrees showed a positive relationship. Xie (2011)  pointed out that incentives stock options have greater long-term advantages in several key aspects of the motivating agent, reducing agent costs and improving corporate governance structure.
Figure 2. Connect the principal’s interests with shareholders by ownership.
2.1.4. Entrenchment Effect
Entrenchment effect is a phenomenon that is too far to get. With the increasing of the agent stake of the company, their controlling and influence on company are increasing, so they can pursuit more personal interests, at the same time it can reduce external constraints, and improve agency costs, resulting in reducing the company’s performance. Bebchuk and Fried (2003)  put forward a theory of management rent extraction, and equity incentive is not considered an effective way to solve the agency problem, but can encourage the agent to use their power to get extraordinary income that can be used to cover rent power and exacerbate the agency problem. Wruck (2004) believe that using a wide range of equity incentives is the reaction of many companies in situation of lacking of effective information and evaluation mechanisms. This indicates that the agent is in the “advantage” status, and this is against the interests of the company.
Other scholars have proposed that equity corporate performance shock did not play a negative role in agency problem, but didn’t play a positive role either, and there is a weak link between the equity corporate performance shock and agency problem. Through statistical regressions analysis which was supported by 1049 companies’ data in1986, Jenson and Murphy (1990) proposed that the relationship between return on equity and the company’s value gains of agents showing a weak trend; Youny and Joung (2001) examined the relationship between company performance and the level of equity incentive, and believe that equity-based incentive arrangements do not increase the company’s performance, and in the case of considering the cost of equity capital and the value of stock options costs, the average performance of the sample companies even are negative. Hall and Knox (2002) confirmed there is lack of a linear relationship between return on equity and business performance of senior management, which makes equity incentive contract vulnerability inside. Through the 2001 annual report of 1036 listed companies to study the relationship between the remuneration structure and corporate performance, Zhen (2003)  found that the operator’s shareholding ratio and operating performance showed a significant weak correlation.
The method described above has different incentives. Modern business case presented a wide range of incentives. Motivation is still formally adopting a new allocation of the business interests and re-creation process. First, based on considering company’s business strategy, operations and culture, combining the agent’s own interests and appetite for risk, both of actors think of realizing two sides’ matching, playing the role of incentives effectively.
3. Separated Monitoring Method
Take monitoring as recording the behavior of managers by the entire board or one of the boards. The aim of monitoring is to ensure agents’ behavior can be taken as needed.
Supervision and restraint mechanism are results of monitoring operators’ different behavior which is based on different business performance, so that principal can do some related punishment. Sometimes, incentives are deformed just because of the failure of the agent oversight mechanisms. Cui  (2006) considered incentives and restraint are interdependent, interacting, and incentive itself does not absolutely prevent opportunistic behavior of the agent, therefore at the same time of improving the incentive mechanism, we must pay attention to maintaining motivation and restraint relative, and strengthening supervision and restraint mechanisms.
3.1. Internal Control: Monitoring That Is Carried by the Board of Directors or Managers
Internal supervision is a variety of interrelated constraints systems approaches for an enterprise to protect the security of their assets, improve its management efficiency and to reduce risk. Internal controlling is based on the principal-agent relationship, on one hand, reconciling the relationship of principal and agent, and strengthening internal controls will help strengthen the rational allocation of the rights and interests, clear and balance everyone’s responsibilities. On the other hand, the operator, that is agent, due to limited personal energy, put the right down and the other managers have rights to make decisions, thus multi-level agency relationship appears, giving birth to the company’s organizational structure, with its internal supervision and controlling show up.
Corporate governance and internal control are closely linked, if entities want to implement effective internal control, it is necessary to consider whether the corporate governance structure is reasonable or not. Therefore, most scholars study the corporate governance structure from the internal control issues. Corporate governance and internal control are closely linked, and if organizations want to implement effective internal control, it is necessary for them to consider whether the corporate governance structure is reasonable or not. Therefore, most scholars study the corporate governance structure from the internal control issues. Li (2007)  proposed that although the modern company organization shapes company’s mode of separation of powers, including shareholders’ meeting, board of directors and board of supervisors, but in practice there is no real power-balanced mechanisms, especially the current situation of implementing company internal supervision: the legal status of the shareholders’ meeting is weak, the supervisors do not supervise as well as independent directors are “not independent”; Xiang (1997)  paid attention to a special behavior of “agents”-the issues of corruption and the effectiveness restricting function of its various corporate governance structure.
Therefore, different scholars put forward their views on how do internal control, for example, on the basis of cleaning the listed companies’ economic ties and contractual relationship, Li Wenchang proposed that it is necessary to establish supervision mechanisms from a board of supervisors, internal audit and other oversight mechanisms, so that the monitoring center can be controlled by high level managers and the operating efficiency of internal controls can be improved; Tian (2008)  believes that if the requirements that management decision-making power can separate from supervision can be achieved, Principal can separate shareholders’ powers and improve agency management, monitor the separation of powers of its shareholders to improve agency management through perfecting monitoring from independent directors Under the circumstances of combining theory with practice, scholar proposed the relevant supervision modes. Tian (2002)  presented that stakeholders take participation in corporate decision-making and implementation supervision equally, at the same time, as an independent supervision actor, every stakeholder does supervision through the corresponding signal display mechanism and proxy voting mechanism. Fu (2010)  summarized three supervision models: the one- tier system board oversight which focus on external oversight: Shareholders’ Meeting―Board of Directors― Executive Committee (managers); two-tier system of general meeting of shareholders: supervisory board― board of directors board of supervisors, and supervisory board is in high position responsible for major oversight in high position; and triangle system which assigned supervision power to the business of the Board, dedicated execution supervisory power to the Board of Supervisors.
3.2. External Monitoring: Monitored by the Controlling Shareholders or the Market That Controlled by Large Companies
Market shoulders a responsible for the task to separate business ownership and operation rights, at the same time, market is a contradiction between principal and agent regulators, so it should focus on strengthening constraints over capital markets and entrepreneur market.
Market mechanisms commence competitive constraint in three parts: agents markets (mostly managers), capital markets and product markets.
First, perfect manager market, establishes competitive match for managers market and in this market, develop performance evaluation, ranking and other dynamic evaluation of effective measures for agents to carry out mo- nitoring activities. Fama (1983)  proposed that making use of market reputation agent urges the agent to work hard. He considered that agent market constraint agents’ behavior, increase the cost of non-compliance, which is an early reputation in the market mechanism. Holmstrom (1991)  established reputation model agents and directly prove that the reputation can be a substitute for explicit incentive contracts, so open ground for research of how reputation affects business incentive problem. According to Varian, another force that constraints agency problem is to use the potential force of competion, so that the mechanism of mutual restraint between the agent is formed. Chen (2014)  presented that to implement comparative performance evaluation to agent (evaluation mechanisms) puts pressure on professional managers through the “survival of the fittest” mechanism in manager market.
It is the competition that attracts more people go into the managers market to compete for the position, and the principal managers can freely seek the right people. On the other hand, it can put a strong pressure on serving managers.
But years later, Holmstrom (1993)  further suggested that reputation mechanism cannot solve all the problems of the agent, at the same time, it will also bring new agency problems. Gilson (1990)  studied 111 listed companies between the bankruptcy reorganization in the United States in 1979-1985, found that bankruptcy makes the banks to become the largest shareholder in most companies, and such banks tend to reappoint board members. After the agent was forced to leave, the opportunity to be a director in other companies again is greatly reduced, indicating that the reputation of directors was severely affected. Holmstrom also found that all of the directors of the target company found it difficult to keep their directorships after taking over. In the future, the availability of outside directors depends on their performance before taking over directorship, the directors who rejected an offer to take over in the poor performance of the company will end their career in the labor market, who accepted an offer to take over do not.
Second, improving the capital markets (mainly stock market) makes share prices reflect business conditions and managers’ abilities. Through the struggle for control of the enterprise, the enterprise’s managers feel the pressure, and thus bound managers behavior. Li (2007)  believes business situation and the stock prices are positively correlated hook. Under the constrained by capital and stock markets and based on stock market information, investors gain insight into the operation and management of enterprises and form a potent effect on the agent.
Again, improving product market and competitive market, in large part can reflect the effort level of business executives operating capacity directly. The market share of goods and services, sales margins and other indicators directly reflect the agent operating capacity.
In addition to these primary oversight and restraint mechanisms, Tang (2009)  proposed the financial supervision that accounting agency do for state-owned enterprises, the supervision and restraint that bank do for managers debts, laws and regulations, ethics, and so is effective external monitoring mechanism.
Combining scholars’ research and comprehensive view with the actual business development, the main problem for the agency to address is to keep the agent behavior in same direction with pursuing business interests, to reduce agency costs, to improve profitability, and the method which can be used to solve problem can be explained from the perspective of excitation and monitoring.
First, research on how principal encourage agents focuses on binding both interests together to do integrated incentives, then takes different specific methods. One of these methods is to connect agent’s remuneration with the performance of enterprises. Different scholars conclude differently. Some scholars find that compensation and corporate performance are positively correlated. The principal can use this method to encourage the agent positively. However, some scholars have suggested that there is sticky or lag effect between corporate performance and agent’s remuneration. That is to say, the growth of corporate performance and agent’s remuneration is not at the same time. The growth of them shows up in inverted “U” which is low left and high right. The other scholars proposed incentive pay not only cannot drive the company’s performance grow, but increase speculation and encourage them to erode the company and shareholders. The other one is to connect agents’ benefit with shareholders through the ownership of shares. The effect of agents’ holding shares can be divided into “convergence of interests” and “entrenchment effect”. The two effects were demonstrated at both ends of the “U” curve.
Secondly, principal strengthen monitoring for agent to prevent the agent speculation can reduce the company’s interests. They commence in two aspects of internal and external monitoring. Internally, corporate governance structure will be adjusted to shape the separation of powers and strengthen the monitoring role of the board of directors. From the outside, market monitoring is the main means, but the agent market, capital and product markets all bring pressure on the principal to monitor the agent. On one hand, the agent market reputation mechanism brings pressure on agents, makes agents’ reputation and corporate performance linked and prompt agent to work hard; on the other hand, it cannot solve all the problems of the agent. Oppositely, it produces new agency problems. The capital and product markets can reflect operating conditions of agent better, form an effective supervision and restraint on the agent.
But for the current evidence on the effectiveness of each of these governance mechanisms are inadequate. The study for multivariate (meta-analysis) ownership structure and corporate performance analysis shows that “a weak relationship exists between the diverse forms of ownership and performance indicators” (Dalton, et al., 2003). The analysis of other governance mechanisms also proved to be significantly blurred (Bebchuk, Fried 2005). Based on previous analysis, some scholars propose that the reason for the lack of strong evidence study for these mechanisms is that they pay excessive attention on independent research of effectiveness of agency problem solving (Aguilera et al., 2008). In other words, in the past, research paid more attention on the role of a governance mechanism to resolve the agency relationship and the agent problem in order to be able to find a “medicine” to agency problems, but ignores the complementary relationship between the “medicines”. Most researchers focus on the ability to “independently solve agency problems between shareholding managers” of governance mechanisms (Rediker and Seth, 1995  ). Based on several studies, there are number of articles which focus on company ownership and corporate performance relationships. These research works clarify the relationship between a particular type of the ownership and corporate performance as a proof of value of ownership of the management mechanism, but ignores other mechanisms (Hunt 1986, Dalton, 2003, Kang, 1999). With meta-analysis of the neglected relationship between all relevant corporate performance and various types of ownership, the future research should go into the “alternative hypothesis” and “complementary effect”.
Domestic and foreign scholars classified agency problems governance mechanisms from different angles, there are two categories, the most widely accepted mechanisms can be divided into internal and external governance mechanism, and the second can be divided into supervision mechanism of the behavior of the agent and the distribution mechanism which is related to agent decision-making behavior.
The further study is based on the different governance categories which take progressive questions, and in the process of the study these questions can be solved in order. The first question is how to classify governance mechanisms based on the study of a common set of governance? Defining a clear classification for governance mechanisms and do more subsequent research on this foundation is good beginning to solve this kind of problem. The second one is where the difference between alternative and complementary relationship? Do they exist in the internal governance mechanism or between different governance mechanisms? The third one, from the view of the relationship between alternative and complementary, when monitoring combines with incentives, how they work together. The fourth, the relationship between alternative and complementary within the governance mechanism could exist between the set of governance mechanisms at same time? Does it affect the operation effectiveness of governance mechanisms set? Further, enriching relevant research of how the governance mechanism set solves the agency problem.
Key project of Philosophy and Social Science Research in Beijing (09AbJG299), key project of training plan of Beijing young top-notch talent (CIT&TCD201304109), key projects of Innovation and Entrepreneurship of Beijing Wuzi University (0541605843).
 Aguilera, R.C.F.I. (2008) An Organizational Approach to Comparative Corporate Governance: Costs, Contingencies, and Complementarities. Organization Science, 19, 475-492.
 Zuo, J.J. and Tang, Y.J. (2010) Excessive Incentive and Corporate Performance—Research Is Based on Marginal Effect of Diminishing and Chinese Executives of Listed Companies Team. Industrial Economics Research, No. 1, 48-56.
 Li, J.B and Zhao, L.J (2010) An Empirical Study of Senior Management Compensation and Corporate Performance—Based on the SME Board Companies from 2006 to 2008 Panel Data Analysis. Journal of Yunnan University of Finance and Economics, No. 2, 80-86.
 Chen, C.H. and Lai, Y.Z. and Li, M.Y. (2014) Research on Governance Mechanism of Family Business Agent—Based on Agency Theory and the Theory of Embedding. China Human Resources Development, 13, 6-11.