Analysis of the Liquidity Lockdown Through the Contagion Mechanism in the Interbank Market

This paper quantifies the spread of contagion in the banking system as a result of an exogenous shock. The methodology used for the analysis is the agent-based network model, which allows us to take into account the behaviour of institutions under different scenarios. Due to the lack of publicly available data on interbank interconnections, the maximum entropy method and the method described in Hałaj G. and Kok C. (2013), to reconstruct the interbank network based on the available balance sheets of Russian banks at the end of 2019 and for Q3 2020. In the analysis of the Russian banking system it was concluded that with a small exogenous shock the full interbank network is more resilient, but as the exogenous shock increases, after a certain point a large number of interconnections in the network contribute to the spread of risks across the network and form a less stable system. Despite the presence of a pandemic factor in 2020, the number of defaulted banks and the rate of contagion spread are comparable between the time-sampled data for 2019 and Q3 2020. The hypotheses that rising NPLs in the real sector, the increasing share of deposits that depositors withdraw early from a bank when its financial position deteriorates; and an increasing discount on the market value of assets are also confirmed to have a negative impact on the stability of the banking system. An important factor is the period over which contagion can occur. Even a small increase in the probability of default leads to an increase in the share of real sector NPLs with some time lag. If during this period (up to 7 months) no anti-crisis changes take place in the banking system, the result may be contagion of a larger part of institutions. When analysing the results of the model in terms of bank classifications, it was concluded that banks under sanation are the first to go bankrupt. The most resilient banks are those with foreign capital, while small and medium-sized banks prevail in terms of number of defaults.