Abstract The study analyzes the influence of geopolitical risk on banking stability in 43 countries and its moderating effect of governance quality, for SDG 8 (Decent Work and Economic Growth) and SDG 16 (Peace, Justice and Strong Institutions). The banking system is vital for economic growth, and in times of greater global uncertainty, it is imperative that the banking sector is stable. We explore the heterogeneous impact of geopolitical risk (GPRHC) on banking stability, using monthly panel data from 2012m1 to 2021m1 as our baseline and additional annual data (2012–2021) for robustness analysis, under various institutional setups. The empirical model features Quantile Method of Moments Regression (MMQR) and Quantile Local Projections (QLP) which allow to address distributional heterogeneity and dynamic responses at various quantiles of banking stability. This is further validated by annual-frequency estimations, bootstrapped quantile regressions and Simultaneous Quantile Regression (SQR). The models adjust for the most important macro-financial variables: financial depth, banking efficiency, risk provisioning, real GDP per capita, inflation, and governance quality, as well as the interaction between geopolitical risk and the latter. The results indicate that the impact of geopolitical risk is asymmetric and differential along the banking stability distribution. The destabilizing effects are higher for more vulnerable banks, especially in terms of credit risk, and adjustment capacity is higher for more resilient banks. Governance quality as a moderating factor helps lower geopolitical shock impacts. The impact of geopolitical risk is further dampened by a strong institutional quality, as demonstrated by a formal marginal effects analysis, particularly for vulnerable banks. In summary, the results indicate that the impact of geopolitical risk on banking stability is contingent on institutional strength and that better governance is crucial to foster greater financial stability and contribute to sustainable development.
Sabuj et al. (Wed,) studied this question.