The use of accounting and stock market data to predict bank financial distress: the case of East Asian banks

This paper investigates whether market information could add to accounting information in the prediction of bank financial distress in Asia. A stepwise logit model is first estimated to isolate the optimal set of accounting indicators and then extended to include market indicators. Dummy variables are also introduced in the model to account for the possible existence of balance sheet structure effects. Our results show that market indicators bring in additional information in the prediction process and this contribution holds whatever the importance of the ratio of market funded liabilities over total assets. We also find that market indicators are significant to predict banks' financial distress whatever assets structure. However, for non traditional banks, that is for banks with a low ratio of net loans to total assets, market information seems difficult to interpret.