基于前景理论的商业银行信贷决策及绩效评价研究

The credit decision made by commercial banks are susceptible to some subjective factors,which would lead to the deviation of the credit decisions.In reality,credit is always over-concentrated in some large enterprises or industries. Firstly, we built a credit decision-making model of commercial banks, introduce some endogenous variables into the models, such as the decision-makers subjective cognitive ability and lending psychological preferences,research the degree of rationality, then find out key factors which impact the decision-making of commercial banks and its influence degree.Secondly,the process of decision-making was divided into two steps:editing and evaluating.At the editing stage, we research on how the index of performance appraisal affect credit decisoin-making based on the prospect theory, using models to prove under constraints of the current performance index, credit decision-makers will be lead to irrational behavior such as herding.The index of performance appraisal decides the “decision-making frame”, so it impacts the predilection of decision-makers then impact the final decision.Then a feeback mechanism was built,the index of performance appraisal was revised in order to weaken the decision-makers irrational behavior.At the evaluating stage,a revised prospect theory effect model was built through weights and value function,and the personal value function identity and decision weights distribution identity were known by decision-maker throught system of derivative function,so it can make the decision more rational.

商业银行信贷决策易受某些主观因素影响造成信贷决策偏差。本研究从银行信贷过度集中于某些行业或大企业等非理性现象出发，首先构建商业银行信贷决策模型，将决策者的主观认识能力、放贷心理偏好等作为内生变量引入模型，考察实际决策时有限理性的程度，找出影响银行信贷决策行为的关键因素及影响程度。其次将决策过程分为编辑和评价两个阶段，其中“编辑”阶段从前景理论角度研究绩效评价指标对商业银行信贷决策的影响，通过模型证明在目前的绩效考核指标下会导致损失规避等非理性行为，不同的绩效评价指标决定不同的“决策框架”从而影响偏好及信贷决策，建立反馈机制，在其作用下不断调整绩效评价指标框架以减轻决策者非理性行为。“评价”阶段研究从权值和价值函数两方面构建修正后的展望理论效应模型，通过开发测试决策者价值函数和权重函数的系统引出函数，使决策者了解自己价值函数特性以及决策权重的分布特征，使自己的决策更加理性化。

本研究从银行信贷市场中传统的期望效用理论不能解释的异像出发，以“前景理论”为基础，通过调查问卷、实证分析、结合模糊数学等方法，对我国商业银行信贷决策及绩效评价问题进行研究，研究发现：（1）前景理论在描述我国商业银行贷款经理人的决策行为时优于期望效用理论，在一定程度上克服了期望效用理论的不足。（2）前景理论与期望效用理论的最大差异，在于不同事件结果的价值函数和相应的权重函数两个方面。当某一信贷配置方案相对于参考点存在收益时，贷款经理人是风险规避的；而当方案相对于参考点存在损失时，贷款经理人是风险寻求的。（3）将前景理论和模糊数学结合的模糊决策方法，模拟商业银行客户经理的信贷决策行为是可行和理的，在对客户的准则值进行评价时，用区间数等模糊形式来表示决策信息，可以更好地刻画信贷决策问题，帮助银行贷款客户经理提高决策质量。（4）对银行考核指标所占权重做些许改动，如减少考核指标中利润目标计划计算完成率所占权重，增加收入结构指标计划完成率和新业务及重点推广业务计划完成率权重，采用改动后的考核标准对各方案的前景值进行计算，发现客户经理选择的最优方案会发生变化。