Efficiency in the Highly Market-Segmented Chinese Banking Sector: A Meta-Frontier Non-Radial Directional Distance Function Approach
33 Pages Posted: 17 Feb 2015
There are 3 versions of this paper
Efficiency in the Highly Market-Segmented Chinese Banking Sector: A Meta-Frontier Non-Radial Directional Distance Function Approach
Efficiency in the Highly Market-Segmented Chinese Banking Sector: A Meta-Frontier Non-Radial Directional Distance Function Approach
Efficiency in the Highly Market-Segmented Chinese Banking Sector: A Meta-Frontier Non-Radial Directional Distance Function Approach
Date Written: February 13, 2015
Abstract
In this paper, we apply the so-called meta-frontier non-radial directional distance function approach to the performance analysis of 143 Chinese banks over the period 2006–13. We find that relative to the group frontier in each market segment, the Big-5 Chinese banks exhibit good efficiency. However, their low meta-technology ratios indicate that they are currently unable to make full use of their technological potential to achieve higher economic goals such as cost-cutting and full profit-maximization. As befits their ownership structure, the joint stock commercial banks display the best technical efficiency. Further, the low efficiency scores and high meta-technology ratios demonstrate that the city commercial banks may serve as the primary implementation channels of local policies, particularly those concerning fixed investment. While still at the start-up stage in China, foreign banks display the highest meta-technology ratios, suggesting they possess the best bank operating environments. Finally, over the sample period, Chinese banks in general exhibit a stable and increasing level of technical efficiency, save a very small decline during the 2008 global financial crisis, whereas cost and profit efficiency exhibit significant fluctuation.
Keywords: Meta frontier analysis, Non-radial directional distance function, banks, China, determinants of efficiency
JEL Classification: C23, D24, G21
Suggested Citation: Suggested Citation