双边信息不对称的供应链契约机制效率:基于实验研究

The Contract Mechanism Efficiency of Supply Chain with Bilateral Asymmetric Information: A laboratory Investigation

契约机制是协调供应链企业行为的重要理论方法,却很少被实验验证。本文通过实验室实验的方法验证双边不对称信息下的供应链契约机制的效率。首先,简化了制造商和销售商双方各自成本信息不对称的数学模型,并基于此给出相应的契约机制和相应的数值算例,以此作为实验设计的基础。其次,按照实验经济学基本规范设计实验并进行了20轮有效实验。实验结果显示:(1)随着实验进行,被试真实共享信息的频率有逐渐增大的趋势,并在实验后半段频率趋于稳定。全程实验中,制销角色双方在真实共享信息比例都在63%以上.(2)被试的个体收益会受到对方信息是否真实共享的影响。如果任一方或双方谎报信息,双方收益都会受损。通过实验,我们发现如果决策者理性进行决策---最大化其期望收益,则会真实的共享信息。因此,理论契约机制在激励制销双方真实共享信息上总体是有效的,其效率损失的主要原因是被试个体的非理性行为。

The contract mechanisms for supply chains are often used to coordinate individual firms＇ actions, and have been extensively studied by scholars for decades. Some contracts are proven efficient theoretically, such as buyback contract and revenue-sharing contract. However, in the experiments they do not work well as theoretical prediction. Thus, the contract mechanism for supply chain with asymmetric information may also have lower efficiency. This paper investigates the efficiency of contract mechanisms for a supply chain with asymmetric information, especially for the bilateral asymmetric case. In the first part, we present a simple model derived from the research on bilateral asymmetric information. First, we simplify the model without considering inventories and lost sales. Second, we propose the coordination mechanism （an ex ante contmc0 that specifies the commodity trading quantity and transfer payments. The transfer payments consist of an incentive scheme, based on the improved AGV （d＇Asprernont and Gerard-Varet） mechanism, and an allocation rule, based on the proportions of information rents. Last, we give a numerical example to illustrate the efficiency of contract mechanisms. In the second part, in an attempt to investigate the efficiency of contract mechanisms, we design experiments, including scene designing, reward designing and error controlling. As most laboratory experiments, we ask students participants to act as experimental subjects. We firstly set an imitation scene in which subjects can make their decisions under the direction of experimental specification. Second, subjects acting as the roles of manufacturer and retailer randomly choose to share their private information, determine their commodity trading quantity and transfer payments. When they get their private profit, they enter the next round and repeat the similar decision-making process again. After finishing all the experiments, subjects claim their rewards. In the third part, we present experimental and analysis results. The experiment consists of a worm-up experiment （2 rounds） and a formal experiment （20 rounds）. Each round lasts about 5 minutes, each one in the experiment received about 55 RMB averagely. The experiment data show that in the early stage subjects share their information opportunistically and in the latter stage they share their true cost information in a large extent. The frequency of sharing information truthfully becomes relatively higher with the experiment＇s proceeds. Overall, the ratio of sharing information truthfully by the manufacturer and the retailer accounts for more than 60 percent. Besides, a win-win situation is created for manufacturers and retailers when they share their private information truthfully, whereas their profit will be damaged if anyone of them shares information falsely. We find that subjects may share information truthfully if they make the rational decision of maximizing their expected profits. This indicates that the contract mechanism on revealing information truthfully is effective, and the reason for efficiency loss is individual＇s non-rational behavior. In summary, this paper investigates the effectiveness of contract mechanisms by means of experiment. The experimental results show the effectiveness of theoretical contract mechanisms. Our research can bridge the gap between the contract mechanism theory and the practice of supply chain with asymmetric information. Future research can make a deep analysis of decisions＇ behaviors by integrating behavior game with physiological theory.