Markov switching volatility connectedness across international CDS markets

Abstract

We analyze the interconnectedness of sovereign CDS premiums to assess risk spillovers over the period from April 9, 2015, to April 1, 2024, which includes major volatility episodes such as the COVID-19 pandemic and the Russia-Ukraine war. By employing time-varying parameter vector autoregression (TVP-VAR) and Markov-Switching-Dynamic-Regression (MS-DR) models, we investigate how volatility transmits across countries. Our findings reveal that volatility spillovers intensify during high-regime periods, with significant events amplifying interconnectedness among sovereign CDS premiums. Furthermore, developed nations such as the US and UK exhibit lower susceptibility to external shocks, whereas countries like Mexico and South Africa act as net transmitters of volatility. Specifically, South Africa emerges as a key risk transmitter during high-regime periods, while Mexico consistently plays a significant role in risk transmission across both regimes.