Global Financial Market Connectedness Under Crisis: Evidence from TVP-VAR and BSTS Models

Abstract

The aim of this study is to examine the volatility spillover and causal relationships among global financial markets during the COVID-19 pandemic, the Russia-Ukraine war, and the economic and political uncertainty experienced in 2024 and beyond. To this end, Time-Varying Parameter Vector Autoregression (TVP-VAR) analysis was used to model time-varying relationships using daily MSCI index data for the period January 1, 2018, and July 13, 2025, and Bayesian Structural Time Series (BSTS) methods were used to reveal the causal structure. The TVP-VAR analysis results indicate that volatility spillover intensified in the short term, particularly during pandemic and war periods, but that short-term shocks extended into the long term during recent periods of economic and political uncertainty. Furthermore, the EM and World indices are net emitters of information, with the Asia index being the largest recipient of information, followed by the Europe index. On the other hand, it was determined that volatility spillovers during pandemics and wars quickly faded, but became persistent during the recent period of economic and political uncertainty. The findings from the BSTS analyses confirmed the TVP-VAR results. It was determined that the Asia and EM indices, in particular, became more fragile during crisis periods, their confidence intervals widened, and these two markets became more sensitive to uncertainty. These two markets were negatively affected by external shocks.