Research On Incentive Contract Based On Rabin Motivation Fairness

Incentive problem is a major problem of economics, management, and even psychology. When people with the different target and asymmetry information compose a business, incentive problems arise. To solve the incentive problems, economists develope beatiful incentive contract theories which assume that people are purely selfish, and only pursue personal material interests. But the psychology, experimental economics, anthropology and biology have demonstrated that people not only concern their material gain, but also concern others motivation and do like this :'returning the favor, tooth for a tooth'.Rabin modelized the reciprocity motivation as fairness utility function, and make great achievement in behavior game theory.Basing on the classical theory of incentive contracts and considering people's motivation fairness preferences, this paper implants Rabin fair game model to the classical incentive contract theory, amends and expanses the classical theory to the new one, and old one is now a special case in the new model. This dissertation provides a new explanation for the inconsistencies between theory and reality, and provides useful suggestions for the enterprise how to design incentive mechanism.Firstly, this paper analyzes the incentive contract based Rabin motivation fairness model when the agent's action is hidden. When the principal can not observe the aggent's action, moral hazard problem arise. This paper discusses the optimal incentive contract structure based on reciprocity preferences both the symmetric information (reveal action) and the asymmetric information (hidden action) cases. This paper theoretically proves that the principle of sufficient statistics advocated by the classical contract theory is no longer hold when reciprocity preferences exist. There are two situations: Firstly the contract is over-determinate; secondly the contract is incomplete. When the principal can observe agent's action without cost, the optimal incentive contract based on reciprocity should not only reflect the agent's efforts, but also the external random factors, that is say that the contract is over-determinate and agent gets the "pay for luck " .The optimal incentive contract shows that when the external environment is good, the agent with high-reciprocal preferences gets more positive "pay for lucky"; when the external environment is poor, the agent with high-reciprocal preferences gets more negative "pay for lucky". This paper also compares the optimal contract with the asymmetric information and asymmetric information. In particular, this paper points out that the agent's information rent of action may be negative. Even the principal can adopt full supervision on the agent, he should give up full supervision. This shows that the optimal incentive contract is not complete. So it answers this question: why most contracts are not complete. It explains experimental results of Fehr, Klein and Schmidt(FKS).Secondly, this paper analyzes the incentive contract based Rabin motivation fairness model when the agent's fairness preference is hidden. When the principal can not observe the aggent's fair preferences, adverse selection problem arise. Therefore, how to design a mechanism to identify the agent's preferences is an important question. In the case of symmetric information, the principal can observe the agent's true type, so the principal can design different contract for different type of agent, that is say that the optimal contract is dependent on the type of agent. In the case of asymmetric information, the principal can not know the true type of agent. If the two types of agents should be employed, then the principal can not distinguish the agents with different preferences, the optimal contract is mixed. Because the optimal contract mixed, there are always some agents get the information rent by hiding their true types. This conclusion is same as that of Siemens which is based on result fairness. We can get a "pooling theorem", that is: if two agents should be employed, it's impossible to design the difference contract to identify the different preferences of the agent, no matter the preference is result fairness or reciprocity. Because the agent can not be treated different, the optimal contract is the sensitive to external good luck, and is non-sensitive to bad luck, in other words, the agent wage is "rigid". So it gives a new interpretation to involuntary unemployment of labor market. At the same time, this paper proves a "Separating theorem ", that is, if the principal intends to hire only a single type of agent, then the principal can screen and select different agents by contract designing. Simply say, the agent with high reciprocity will help you when you were suffering, the agent with low reciprocity (or purely self-interested agents) will share with you when you make rich. Phenomenon that"work together the first year, be prosperous the second year, and disband the third year"can be explained in economic sense. The principal can screen and select the different agent by using participation constraints and make more expected profit.Third, this paper analyzes the team production based Rabin motivation fairness model. Holmstrom classic team production model assumes that participants have a purely self-interest preferences, and proves such impossibility theorem that'a team can not achieve Pareto optimality with the budget balance'. When members of a team have Rabin fairness preferences, the team's productivity is higher than that the classical model predicts. So the team hires members with Rabin fairness preferences can achieve Pareto improvement and Pareto optimal. Therefore, Holmstrom's theorem that'a team can not achieve Pareto optimality with the budget balance'does not hold when considering Rabin fairness preferences. So it explain the confliction between the theory and reality.Finaly, this paper analyzes the bargaining problem when making a incentive contract. The traditional theory assumes that the principal offer a "take it or leave" contract to the agent, and puts aside bargaining problem. If this assumption is true, then the principal can get the most revenue, and the agent can only get reserve revenue. This is extremely unfair. Therefore, the agent may reject the contract, and offer counter-contract. If the principal refusal will bring about the loss, then the principal's"accept it or leave" contract is incredible. Therefore, the principal and the agent may make Rubinstein bargaining. The counter-contract offered by the agent meet the principal participation constraint, and maximizes the agent's certainty equivalent income. However the principal can only get their own reserved earnings. If the principal is risk-neutral, the agent is risk aversion, and the agent's efforts can not be observed, then the traditional contract can not be achieved neither optimal risk-sharing nor the optimal level of effort. However the counter-contract can simultaneously achieve the optimal risk-sharing and the optimal level of effort.When the principal and agent make incentive contract by Rubinstein bargaining, the period length doesn't affect agent's share of a unit of output and effort level, only affects the agent's fixed-income. Bargaining process increases participation constraint. The agent gets only reserved income in traditional contract, while gets more when bargaining. So contract with bargaining makes the principal and agent share the total surplus, and is fairer than traditional contract.