Cost efficiency analysis in banking industries of ten Asian countries and regions

Abstract

Despite the great achievement of three decades of economic reform, the Chinese banking sector takes the blame for its dysfunctional system, especially the large amount of non-performing loans. The ease of foreign banks’ entry set by the WTO from December 2007 raises our concern of the capability of domestic banks to compete against foreign Asian banks. This study attempts to address this issue by measuring the cost efficiency of ten major Asian banking industries from 1998 to 2005 using panel data stochastic frontier approaches. Based on our preferred consistent panel data estimating models, the higher cost efficiency score from including cross-country environmental variables suggests that differences between countries can explain part of the inefficiency. We also find that the overall cost efficiency level of Chinese commercial banks ranks in the fifth place, suggesting that Chinese banks still need to strengthen their ability in competition. Some policy implications are also suggested.

Keywords:

• stochastic frontier approach
• panel data
• banking efficiency

JEL Classifications:

• C23
• D24
• G21

Acknowledgements

The authors are grateful for helpful comments from referees and participants in the 19th Annual Conference of CEA (UK) in Cambridge 2008, North American Productivity Workshop in New York 2008 and International Workshop on China's Productivity and Joint CEA Annual Conference in Hangzhou 2008. Any remaining errors are the responsibility of the authors.

Notes

Notes

1.  RMB is Chinese currency, Ren Min Bi.

2.  To save space, the review and references for the survey of 15 cross-country studies are not provided in this paper but are available from the authors upon request.

3.  Although the cross-country heterogeneities are time-varying over the sample period, these differences are trivial and one cannot rule out the time-invariant characteristics in these time-varying heterogeneities. Moreover, based on the likelihood ratio test, it is suggested that treating these time-varying heterogeneities as time-invariant, estimates of the parameters and inefficiency scores are quite similar, which suggests that the time-invariant heterogeneities are the latent characteristics in the cross-country heterogeneities.

4.  The Hausman test favours the FE rather than RE, which often occurs in panel data models. However, we can still justify the use of the RE model. The Hausman test is based on the classical panel data models and concentrates on whether the individual-specific effect is random- or fixed-effects. But in an efficiency study, what we are really interested in is whether the inefficiency is best modelled by random or fixed effects. The Hausman test would simply reject the former based on the existence of correlatedness between the individual-specific effects and the regressors, but not necessarily the inefficiency itself. Since the inefficiency term in the fixed-effects model might capture all the time-invariant heterogeneity, there is a possibility of correlatedness between this heterogeneity and the regressors. Ultimately, however, we wish to focus on the inefficiency term, which actually is not correlated with the regressors.

5.  We are unable to compare our results with others due to absence of Asian banking efficiency studies in the sample period.