<![CDATA[Background: Despite the growing importance of stress testing as a supervisory tool, its role in reducing information opacity and improving credit risk management remains underexplored in ASEAN’s dual banking system, where Islamic and conventional banks operate under distinct financial structures. Method: Using a Panel Vector Autoregression (PVAR) model with Generalized Method of Moments (GMM), this study analyses 72 listed banks across five ASEAN countries over 2012–2022. Credit risk is proxied by NPL/NPF ratios, incorporating key macroeconomic and bank-specific variables. Results: The findings indicate that credit risk dynamics are highly persistent, with own shocks accounting for more than 97% of forecast error variance in both banking systems. GDP growth and real interest rates emerge as the most influential macroeconomic determinants. Islamic banks display mean-reverting credit risk behaviour, whereas conventional banks exhibit greater persistence. Impulse response analysis reveals that macroeconomic shocks have statistically significant but heterogeneous effects across bank types. In addition, Granger causality results suggest that macroeconomic variables can serve as early warning indicators of credit risk. The COVID-19 period provides additional evidence that stronger capital buffers help mitigate the transmission of macroeconomic shocks to NPL ratios. Conclusion: These results support the role of stress testing as a tool for improving risk assessment and strengthening supervisory oversight in ASEAN banking systems. This study contributes by providing the first PVAR-based comparative analysis of stress testing in ASEAN’s dual banking system, incorporating the COVID-19 shock as a natural experiment, and offering cross-country evidence on the role of stress testing in improving financial transparency.]]>
Copyrights © 2026
