Systemic Soundness and Banking Regulation

Abstract

This study offers a Monte Carlo approach to simulating aggregate default rates and seeks to capture the cross-country variation in these rates with a number of banking regulatory and supervisory practices. The jump-diffusion model produces default rates that out-perform the alternative proxies for systemic risk, including market proxies such as asset or equity prices and volatilities, leverage, as well as aggregate default rates as derived from the Merton, KMV, and random-walk models. Also, the jump-diffusion model captures both 'normal' and 'abnormal' vibrations in the asset price and thus provides accurate and stable aggregate default rates. The proposed model could be relevant for supervisory-review work on procyclicality risk and stress-testing under Basel II. The econometric results show that private monitoring serves as a robust disciplinary tool for mitigating systemic risk. In addition, these results offer partial support for a number of bank regulatory and supervisory practices such as guidelines for bank diversification, restrictions on bank engagement in mutual-fund activities, and supervisory-review actions. While some regressions shed light on the potential trade-off between private monitoring and supervisory oversight, this trade-off is not big enough to offset the beneficial effects of the new Basel II pillars. In summary, 33 countries are likely to face a fall in aggregate default frequency for a given improvement in private monitoring, and 27 countries are likely to face a fall in aggregate default frequency for a given improvement in supervisory power. Thereby, there is ample room for optimising the mix of the Basel II pillars to enhance the systemic soundness of at least two-thirds of the sample countries.