中国商业银行整体效率研究——基于具有中间投入和中间产出的综合网络DEA模型

THE STUDY ON CHINESE COMMERCIAL BANKS OVERALL EFFICIENCY——Basing on Composite Network DEA Model with Intermediate Input and Intermediate Output

本文基于对银行业务结构和特征的分析,将银行看做具有中间投入和中间产出的串联两阶段网络生产结构。第一阶段为负债业务和中间业务,第二阶段为资产业务。基于这一网络结构,本文首先建立具有中间投入和中间产出的综合网络DEA模型,并证明在此模型下整体网络DEA有效等价于每个子阶段DEA有效,即此模型是打开黑箱的网络DEA模型,并定义网络决策单元的投影。本文以中国16家上市商业银行为样本,测算了2006—2010年各商业银行的整体效率、子阶段效率、关联指数和每家银行的投影。分析结果显示,大型国有商业银行在负债业务和中间业务阶段具有明显的竞争优势,股份制商业银行在资产业务阶段具有效率优势,城市商业银行的两阶段业务关联性最差。本文计算的投影,给出了银行提高整体效率的改进方向。

Basing on the deeply analysis, we regard the banking business as a two-stage network with intermediate input and output. The first sub-process is characterized by liabilities and intermediary business. The second sub-process is characterized by assets. Based on the network, we construct a network DEA model with intermediate input and output. Moreover, we prove that network efficiency is equivalent to sub-stage efficiency, which means our model can reveal the inner structure of the black box. We also calculate the overall efficiency, sub-stage efficiency and correlation index of 16 Chinese listed commercial banks from 2006 to 2010. The results show that state-owned commercial banks revealed obvious advantage in the liability and intermediate business. Joint-stock commercial banks are most efficient in assets busi- ness. And city commercial banks have the lowest correlation index.