| Abstract | This study contributes to the well-established efficiency literature with respect to transition economies and developing counties. Although bank efficiency has been a
popular research area in both developed countries and less developed nations, it has been scarce in China mainly due to the lack of data. This study is considered as the first study that comprehensively investigates bank performance using multiple methodologies of financial ratio analysis and stochastic frontier analysis for the period of 1995-2005. The effects of a variety of governance changes have also been differentiated in order to providing policy recommendations for the on-going banking reform. Meanwhile, this study has addressed a number of methodological issues and has developed a more comprehensive stochastic distance function model by combining advantages of existing models, approaches, methods and procedures.
Having experienced fundamental banking reforms for more than a quarter of century, the Chinese banking System has stridden towards a modem banking System with significant improvements in profitability, capitalization, and assets quality. Despite of these observed improvements, the banking system is still associated with relatively
low profitability and capitalization, poor asset quality, and less liquidity, when benchmarking to 7 selected international renowned banks. One of the most impressive progresses has been the significant decrease in both outstanding NPLs balance and NPL ratio. However, it has noticed that the threat of NPLs problem to the economy as a whole remains unsolved.
This study has rationalized economic foundations for the banking reform in China being the principal-agent theory and the budgetary constraint theory. The performance of Chinese banks has been improved and the estimated effîciency level is consistently at 75% in terms of technical efficiency, cost efficiency and profit efficiency. Employing the method of Berger et al. (2005), this study has jointly analyzed the static, selection and dynamic effects of governance changes. Joint-stock
ownership has resulted in outstanding performance, while state ownership has been associated with low technical efficiency and profit efficiency but high cost efficiency.
Foreign banks are more profit efficient but less cost and technical efficient (static effects). Foreign investors have rationally made their investment decisions by
selecting more cost and technical efficient domestic banks, while less profitable domestic banks have been chosen for going public in line with government intension of reforming the unprofitable SOCBs (selection effects).
Attracting foreign strategic investors and encouraging banks going public are two major partial privatization strategies, which have been generally proved as effective
reform measures. The former tends to have positive impacts on technical efficiency and cost efficiency, while significant short-term gains in profit efficiency have faded in the long-term. The expected profit advantage of foreign ownership seems to take an even longer time to be realized. Going public strategy has resulted in performance
improvement in the long-term after short-term losses (dynamic effects).
We can not form a conclusion on whether the reform has succeeded, while what we can conclude is the reform is on the track with right direction. It is important to
construct good corporate governance, but it is more important to ensure the good governance functioning. If those deep-rooted problems, such as government
intervention and NPL problem, can not be dealt with properly in the near future, the chance of success is very small. Thus, our policy recommendations include
consolidating up-to-date reform achievements, improving bank's managerial and operational skills, and reducing state's share in banks to lessen government interventions.
Estimated efficiency is found to be sensitive to the differences in the definitions of outputs and inputs, especially in the presence of high level of NPLs. The income-based model is superior to the earning assets-based model in the estimation of technical efficiency. Similarly, profit efficiency appears to be more appropriate performance measures over cost efficiency. However, we suggest the use of multiple models and measures to reveal more valuable information. Moreover, in estimating cost function and alternative profit function, market average input prices are found to be more appropriate than banks' specific input prices. |
|---|