Essay 1
This paper explores the efficiency of banks and its determinants in the South East Asian region using the DEA technique and Tobit regression. Efficiency was found to decline significantly between 1998 and 2004, suggesting that the effects of deregulation were slow to materialise. Being consistent with comparative studies in emerging markets, foreign banks appear to be more efficient than their domestic counterparts but, conflicting with the general expectation; state-owned banks are less inefficient than private players. Banks with higher levels of government ownership are more technically efficient while those with greater private stakes obtained lower levels of technical (and cost) efficiency. Among
country-level factors, national banking sector development is found to have a strong and positive link with bank efficiency. The results are robust to different modelling specifications.
Essay 2
This paper examines the motivations for foreign bank entry into South East Asian
countries in the aftermath of the 1997 financial crisis. The results show that manufacturing FDI and bilateral trade exert a weak impact on the decision of entry by foreign banks, providing little evidence for the argument that banks follow their home customers abroad as suggested by one strand of the literature. In contrast, local profit opportunities appear to be the prominent factor attracting foreign bank penetration in South East Asia during the
period 1998 to 2004. The results are robust to different modelling techniques.
Essay 3
This paper investigates the effects of competition on risk-taking behaviour at the bank level in South East Asia. The Panzar and Rosse (1987) H-statistic is used as a measure of banking competition for a study of commercial banks from a sample of four countries in South East Asia (Indonesia, Malaysia, Philippines and Vietnam) and we show that it is not necessary for policy makers to increase bank systemic risk in return for a more competitive banking system. In contrast, the results reveal that competition helps to decrease instability. Our results are robust to alternative risk indicators, different Hstatistic modelling and specifications.