Modeling and Estimating the Risk of the Banking System in Form of a Network Model Using CoVaR

​In this paper, a multi-layered network model of the banking system is designed to explain the systemic risk of the banking system of Iran. This model shows how the dependence of the balance sheet structure of banks results in the crisis from one bank to another, and ultimately causes a crisis in the economy as a whole. It is assumed that the banking system is composed of banks with their balance sheet structure dependent on each other. To estimate the systemic risk of the banking system, the daily index data of banks indexes are used between December 2008 and April 2018 and the value at risk of returning daily index data is estimated using an exponential GARCH model. The daily returns of the stock market index are taken as representative of the real economy, Quantile regression is estimated at two levels of 50% and 1%. In the following, based on the estimated parameters in quantum regression and also on the basis of the CoVar Adrian and Brunnermeier (2016), the risk of the banking system is estimated. The results show that the average ΔCoVaR is estimated at -0.8587, which is consistent with negative expectations and represents a high systemic risk of the banking system.