Dynamic interconnections and spillover among sovereign – bank nexus in Bangladesh: a TVP-VAR analysis
Shanza Maryam Khan, Shahzad AkhtarPurpose
This study aims to examine the dynamic interconnections among sovereign – bank nexus in Bangladesh, including sovereign risk, credit risk, liquidity risk, interest rate risk, risk taking behavior and macroeconomic indicators gross domestic product (GDP) growth, Inflation rate and unemployment rate and their mutual influence. Further the time-varying parameter vector auto regression (TVP-vector auto regression (VAR)) framework has been implied to examines the sovereign – bank nexus.
Design/methodology/approach
The research uses a data set of 58 Bangladeshi banks from 2008 to 2024 (718 observations) and uses the TVP-VAR model with Bayesian Markov Chain Monte Carlo (MCMC) estimation.
Findings
The findings of this study, focuses on the TVP-VAR connectedness analysis of 58 Bangladeshi banks between 2008 and 2024, shows that the spillover index for all banks in the country is 70.16%, which shows high interconnection between sovereign risk, banking sector variables and macroeconomic variables. The debt-to-GDP ratio (D/GDP) is a net transmitter of risk (to other value 86.18), connected with moderate credit risk (debt-to-equity ratio 13.45), risk taking behavior (NPL 12.19), liquidity risk (current ratio 10.54) and interest rate risk (net interest margin [NIM] 11.13). The leading transmitters (in other words, the ones that are net transmitters of shocks) are inflation (to other 89.61, from other 80.44), GDP growth (to other 73.47, from other 73.14) and three banking sector variables: debt-to-equity ratio (from other 74.48), NPL (from other 68.45) and NIM (from other 51.24), are net receiver variables, meaning that they absorb shocks rather than amplify them. All the connectedness values are in the moderate range (5%–15%) indicating that the indirect transmission channels mainly interest rate and exchange rate channels dominate the direct transmission channel of sovereign default concerns, and that sufficient bank liquidity and capital buffers reduce the sovereign – bank nexus in Bangladesh relative to the European “doom loop.”
Originality/value
The novelty of this study is the novel application of the TVP-VAR framework to study Bangladesh’s sovereign-bank nexus, which is extended to the Bayesian MCMC estimation method to overcome the shortcomings of rolling-window VAR and dynamic conditional correlation generalized autoregressive conditional heteroskedasticity models, as the new model captures the direction, magnitude and persistence of spillovers over time. The results, which differ from the European “doom loop” theory, show that sovereign risk is mainly transmitted via indirect channels, such as interest rates and foreign exchange rates, instead of direct fears of default and that the banking sector in Bangladesh is a shock absorber, not an amplifier. Furthermore, GDP growth volatility and inflation have been identified as the dominant net transmitters, which puts a new light on the theoretical hierarchy of the net transmitters of systemic risk in emerging economies. These contributions help further the literature on systemic risk and financial contagion and give evidence-based advice to developing countries on issues of debt sustainability, sovereign exposure limits and macrofinancial coordination in countries with incomplete implementation of Basel III.