Volatility spillovers during times of crisis : measuring connectedness between the largest US, UK and European banks

Abstract

This study focuses on measuring volatility spillover and connectedness between the largest banks in the US, the UK and Europe. The research looks at several aspects of connectedness, varying from systemwide, pairwise and directional connectedness. Furthermore, the study looks at the dynamic nature of volatility connectedness to identify varying trends between tranquil time and times of crisis. The research is based on the spillover index framework developed by Diebold and Yilmaz (2011) and utilises daily stock data for the sample banks for the period between 2007 and 2020. The results outline high overall connectedness levels at 83.7%. The rolling sample outlines that connectedness peaks in times of crisis. By splitting the data sample into different time periods, the results show that connectedness between the selected banks remained high at all sub-samples, including periods that did not include any significant crisis. The research further outlines that pairwise connectedness tends to be higher between banks operating in the same jurisdiction. Furthermore, the results show that bank size can be a determining factor in its overall volatility spillover levels onto the system.