The effect of the interbank network structure on contagion and common shocks

Abstract

This paper proposes a dynamic multi-agent model of a banking system with central bank. Banks optimize a portfolio of risky investments and riskless excess reserves according to their risk, return, and liquidity preferences. They are linked via interbank loans and face stochastic deposit supply. Comparing different interbank network structures, it is shown that money-center networks are more stable than random networks. Evidence is provided that the central bank stabilizes interbank markets in the short run only. Systemic risk via contagion is compared with common shocks and it is shown that both forms of systemic risk require different optimal policy responses.

Interbank loans, collateral and modern monetary policy

“…calibrated on a sample of 1549 German banks in 2013, with low dispersion values to limit possible distortions. Risk aversion for the short-term assets is set according to Georg (2013). The scaling transformation of the risk-aversion on the long-term assets is chosen simple, but it has no effect on the aggregate dynamics of the results.…”

“…Risk aversion is divided into two groups, depending on investments' maturity: γ s reflect the risk aversion towards short-term assets and γ l towards long-term assets. Following Georg (2013), the former is uniformly drawn from [γ min , γ max ] and the latter is the maturity-dependent multiple of the former and takes the form γ l = τ γ s /2. Accordingly, Σ is the variance-covariance matrix between the returns.…”

“…They also argue that heterogeneity across banks leads to larger direct and indirect contagion effects, making the whole network structure more fragile. Georg (2013) studies the influence of various network structures on the shock-resilience in a multi-agent interbank market. He documents that central bank policies are only stabilizing in the short term and he argues that, in case of unsecured transactions, random networks are less stable than their money-center equivalents.…”

Heterogeneous beliefs in over-the-counter markets

“…This finding supports previous work in this area. Both Ladley and Bullock (2008) and Wilhite (2001) find that small world type short cut connections can reduce search costs and make the price formation process quicker, whilst Georg (2013) finds that the configuration of an inter-bank market, not just the connectivity, is important in determining the susceptibility to systemic shocks.…”

“…Several papers have considered the effect of particular network structures e.g. Battiston et al (2012); Georg (2013); Iori et al (2006); Ladley (2013); Lorenz and Battiston (2008) and have shown that the connectivity (the number of links between traders in the network) and the configuration of linkages both play a role in market stability.…”