Destabilizing effects of bank overleveraging on real activity: an analysis based on a threshold MCS-GVAR

We investigate the consequences of overleveraging and the potential for destabilizing effects from financial- and real-sector interactions. In a theoretical framework, we model overleveraging and indicate how a highly leveraged banking system can lead to unstable dynamics and downward spirals. Inspired by Brunnermeier and Sannikov (2014) and Stein (2012), we empirically measure the deviation-from-optimal-leverage for 40 large EU banks. We then use this measure to condition the joint dynamics of credit flows and macroeconomic activity in a large-scale regime change model: A Threshold Mixed-Cross-Section Global Vector Autoregressive (T-MCS-GVAR) model. The regime-switching component of the model aims to make the relationship between credit and real activity dependent on the extent to which the banking system is overleveraged. We find significant nonlinearities as a function of overleverage. When leverage is standing above its equilibrium level, the effect of a deleveraging shocks on credit supply and economic activity are visibly more detrimental than at times of underleveraging.