Macro determinants of nonperforming loans and stress testing application on Vietnam banking sector
Abstract
This study investigates the relationship between several macroeconomic factors and the nonperforming loan ratio in Vietnam banking system by using three panel regression models including Pooled OLS, the Fixed-effect model and the Random-effect model. The study’s sample consists of eight listed banks representing approximately 50% of market share of The Vietnam banking system. The examined period is from the fourth quarter of 2008 to the second quarter of 2013. Consistent with many international and domestic evidences, we found that the GDP growth rate is negatively related with nonperforming loans (NPLs) while the lending rate is positively related with the NPLs. Contrary to other studies, we found no significant impacts of the inflation and the exchange rates on the nonperforming loan ratio of Vietnamese commercial banks.
Using the Pooled regression result, we conduct macro stress testing to predict the levels of the nonperforming loans and the expected losses that banks could suffer. The study applies alternative stress testing approaches including conventional approach and Value-at-risk (VaR) approach. In which, this is the first time that any study using VaR to conduct stress testing in Vietnam. The forecast result shows that the minimum capital requirement for banks to survive the adverse scenarios is about 6% as end of 2014. Therefore, banks need to manage their capital above this level and the regulator need to enforce that banks must have at least this level of capital for credit risk.
Keywords: Nonperforming loans, credit loss, capital adequacy, panel regression, stress testing, vector autoregressive model, value-at-risk.