高声誉企业质量数据造假行为演化博弈分析:神户制钢案例  

Evolutionary Game Analysis on Quality Data Fraud Behavior of High-Reputation Enterprise:A Case Study of Kobe Steel

作　　者：赵磊[1] 毕新华[2] 赵安妮[2] ZHAO Lei;BI Xin-hua;ZHAO An-ni(School of Economics and Management,Jilin Jianzhu University,Changchun 130019,China;School of Business and Management,Jilin University,Changchun 130022,China)

机构地区：[1]吉林建筑大学经济与管理学院,吉林长春130019 [2]吉林大学商学与管理学院,吉林长春130022

出　　处：《数学的实践与认识》2023年第7期61-75,共15页Mathematics in Practice and Theory

摘　　要：揭示高声誉企业组织蜕化过程中声誉效应缺陷的作用机理,对完善声誉治理机制具有重要意义.采用独特的内含型设计,在神户制钢门前因分析的基础上,建立质量数据失真风险输出模型,进行动态演化博弈分析并验证.结果表明:组织蜕化过程分为试探、默许、暴露和内控4个阶段;声誉效应依次发挥声誉激励正向效应、声誉激励负向效应、声誉惩罚效应和声誉惩罚与外部监管的叠加效应.在默许阶段,企业声誉随质量数据失真风险输出水平的增大而增大,声誉激励负向作用下低风险、高声誉企业转变为高风险类型.在内控阶段,声誉惩罚与外部监管的叠加作用下,高声誉企业的隐患损失分配系数远大于最优分配系数,促使其蜕化为具有适应性的质量数据造假者.基于理论模型计算信任拐点、不信任拐点和隐患损失最优分配系数,可确定企业的过程能力上限、质量数据失真风险输出水平及隐患损失承担份额,为企业声誉的量化治理提供依据.It reveals the mechanism of the reputation effect defects in the process of depreciation of high-reputation enterprises,and has an importance to the improvement of reputation governance mechanism.By taking unconventional inset design,based on analysis of antecedents of the Kobe Steel's case,the quality data distortion risk output model was established and the evolutionary game analysis was carried out,finally the law of the reputation effect defects was verified.The results show that the process of organizational degeneration is divided into four stages:trial,acquiescence,exposure and internal control.The reputation effect in turn exerts the positive effect of reputation incentive,the negative effect of reputation incentive,the effect of reputation punishment and the superposition effect of reputation punishment and external supervision.In the acquiescence stage,the reputation of the company increases with the increase of the output level of the quality data distortion risk,and the negative effect of reputation incentives causes the low-risk suppliers to become high-risk types.In the internal control stage,the superposition of reputation punishment and external supervision makes the high-reputation enterprise's losses distribution coefficient is much larger than the optimal distribution coeficient,which makes it degenerate into an adaptive quality data counterfeiter.By calculating the trust inflection point,the untrusted inflection point and the optimal distribution coeficient of losses,the enterprise's process capability cap,quality data distortion risk output level and potential losses share can be determined,which provides a basis for quantification of corporate reputation governance.

关 键 词：质量数据造假 声誉效应 神户制钢 演化博弈 

分 类 号：F416.31[经济管理—产业经济] F406.7F224.32