Research On The Efficiency Of Chinese Large Commercial Banks Based On The Three Stage DEA Model

The financial industry has been the mainstay of the national economy development, the bank has been in financial system occupies a dominant position. Since 2005, traffic bank listed in Hong Kong, Chinese large commercial banks in droves successfully listed lays the foundation for the a new milestone in Chinese banking industry. 2006 Chinese financial sector is fully open to the outside world, in 2008 the United States sub loan crisis and then led to the European debt crisis, making the financial crisis sweeping the globe. The increasingly fierce competition in the domestic banking and the influx of foreign banks and Chinese banks points a cup of soup, the internal and external situation should not be underestimated.The efficiency of commercial banks in China has always been a domestic scholars pay great attention to, the measure of efficiency is of the input and output of the enterprise, is a measure of enterprise institution into a certain amount of resources, to achieve the level of output, efficiency reflects the level of competitiveness of the bank enterprise. Currently domestic and foreign scholars mainly uses non parametric method of data envelopment analysis(DEA) and derived from the traditional DEA model of super efficiency DEA, Two-stage DEA, three stage DEA model to measure the efficiency of banks. In addition to the three stage DEA model, other DEA models do not eliminate the impact of external environmental factors and random error terms on the input indicators.In this paper, the three stage DEA model is used to study the efficiency of 14 large commercial banks in China during 2009-2014. The first stage still take the traditional DEA model in the input oriented BCC model, three input indicators for fixed assets, interest payments, and the number of employees, two output indicators for deposits and loans and pre tax profits(profits). Because of the first phase sample data by environmental variables and random error influence, so in the second stage model, this paper take SFA model parameters, to get rid of environment factors and random error on the selection of input values influence. In the second stage, the three environmental factors, which were selected as reference, were established, and the number of the property right structure and the number of branches. Finally, according to the second stage of the SFA regression results, the original input values were adjusted to the third stage of the BCC model in the DEA model.The result shows that the empirical analysis of the second stage of the SFA model excluding through three kinds of environment factors, adjusted input is the third stage of the results, with the first phase differences do exist, and the efficiency of the third stage results are generally higher than the first stage. It illustrates the necessity for adjustment of input in the second stage of the SFA model, after adjustment for the environmental factors and random errors, the results can more accurately represent the bank’s operating efficiency, help Chinese large commercial banks management according to the specific circumstances of the bank to make up their own shortcomings.