China's Commercial Banks Pro-cyclical Capital Buffers Study Differences

After the Financial Crises, the international banking supervisory system putsmuch more emphasis on the issue of pro-cyclicality. As Basel Accord linked capitalregulation to the cycle of economy, more and more debates have been produced onbanks’ capital adequacy of Basel Accord. With some proof in the pro-cyclicality ofthe capital requirement, some literatures noticed that most bankers keep quitesignificant capital buffers in some cases. The behavior of these buffers and inparticular, their relationship to the business cycle probably thus take an importantpiece of information.The paper analyzes the pro-cyclicality of the regulatory capital buffer of51Chinese local banks in the period2004-2011, with a common partial adjustmentmodel and two-step Generalized Methods of Moments. The estimation result ofgeneral group shows that there is a negative relationship between the buffer capitaland economic cycle, which means banks’ capital buffer has a pro-cyclicality.For sub-groups of banks, distinguishing stated-owned stock-joint commercial,stock-joint commercial and city commercial banks and also large, medium and smallbanks, we find the buffer capital pro-cyclicality heterogeneity of commercial bankswith different sizes and characters, after introducing dummy variables interacted withthe independent variable GDP. When grouped by different sizes, estimation resultsshow that large banks with total assets of more than1,000billion RMBs iscounter-cyclical, while both medium banks with total assets of between1,000billionRMBs and small banks with total assets of less than100billion RMBs arepro-cyclical. Furthermore, small banks are more pro-cyclical than medium banks.When grouped by different characters, the estimation results show that state-ownedjoint-stock commercial banks are counter-cyclical, while joint-stock and citycommercial banks are pro-cyclical, and the latter is more pro-cyclical than the former.The paper carries out a robustness test using Return on Assets (ROA) instead ofReturn on Equity (ROE), and Reserves for Impaired Loans instead ofNon-performance Loans. Fortunately, the results are robust. The end of the papergives some regulatory suggestions about the upcoming Basel Ⅲ and thecounter-cyclical supervisory.