An empirical examination of bank risk taking - Asian evidence

Abstract

An empirical examination of bank risk taking, following the financial crisis of 1997/1998, in ten Asian countries is undertaken. The major motivation is that prior research generally examines banks in developed countries by using capital market risk measures, and very little has been done in the context of developing countries, particularly when few banks are listed in the capital markets, such as in the case of many countries in Asia. The thesis consists of four essays and employs two datasets of Asian banks. The first essay examines the relation between accounting and capital market measures of risk using a sample of 46 listed banks from Hong Kong, Indonesia, Malaysia, Pakistan, Philippines, Singapore, South Korea, Sri Lanka, Taiwan and Thailand during 1998-2003. Using the same dataset, the second essay investigates the presence of bank moral hazard and the disciplining factors of risk taking. The third essay examines the impact of the introduction of deposit insurance (the Blanket Guarantee Scheme-BGS) and the changes in capital regulation on bank risk taking using a sample of 65 Indonesian private banks during 1995-2003. Using the same dataset to the third essay, the fourth essay investigates the relation between ownership structure and bank risk taking along with the impact of the regulatory changes on this relation in Indonesia. A panel data methodology is used throughout this thesis. A standard fixed­ effects and random-effects regression analysis is implemented, and the results from the pooled OLS are also presented for comparison. Data from several sources and databases are collected, and a balanced panel is employed. The results suggest that, in Asian banks, selected accounting ratios are related to capital market risk measures. Therefore, these ratios may be used interchangeably and complementarily with their associated capital market risk measures in the analysis of bank risk taking. This finding is important, as most Asian banks are not publicly listed. Moreover, moral hazard exists, indicating that market and regulatory disciplines are insufficient to prevent banks from taking excessive risk and exploiting the safety nets. Bank self discipline (charter value) appears to play a limited role in mitigating risk taking. Evidence from Indonesian banks suggests that the introduction of deposit insurance in 1998 increases leverage risk, whereas a lower capital requirement during the sub-period 1998-2000 raises leverage risk and liquidity risk. Bank moral hazard exists following the adoption of the BGS, and the extent of moral hazard is stronger under a lower capital requirement. In addition, ownership concentration is positively related to overall risk, but negatively related to credit risk and liquidity risk. When deposit insurance is introduced and capital requirements are changed,the Listed and Insured Banks show a greater reaction than the Unlisted and Insured, and Unlisted and Uninsured Banks. We also document that government ownership concentration significantly affects bank risk taking. In general, the results seem to be driven by selected government intervention and poor economic conditions following the 1997/1998 crisis. These findings are important because Indonesia is typical of Asian countries where the economy and the banking system are highly correlated, and government intervention would be most pronounced during economic crises.