A Network View on Interbank Market Freezes

Abstract

We study the liquidity allocation among European banks around the Lehman insolvency using a novel dataset of all interbank loans settled via the Eurosystem's payment system TARGET2. Following the Lehman insolvency, lenders in the overnight segment become sensitive to counterparty characteristics and banks start hoarding liquidity by shortening the maturity of their interbank lending. This aggregate change in liquidity reallocation is accompanied by a substantial structural change that can best be characterized as a shrinking of the interbank network. Such a change in the network structure is consequential: banks with higher centrality within the network have better access to liquidity and are able to charge larger intermediation spreads. Therefore, we show the existence of a sizeable interbank lending channel.

Network centrality and funding rates in the e-MID interbank market

“…Lenders do not benefit from network centrality, and as such, it could be that the market perception about their network positioning (i.e. fragility) dominates their strategic location for intermediation (as in Gabrieli and Georg, 2014). The regression analysis also highlights that there is heterogeneity across different measures of network centrality on how they affect interbank spreads.…”

“…Third, as argued by Booth et al (2014), financial institutions with more extensive and strategic financial networks acquire and process information more efficiently due to their better access to order flows. Fourth, as stressed by Gabrieli and Georg (2014), banks with higher centrality within the network have better access to liquidity and are able to charge larger intermediation spreads.…”

A dynamic network model of the unsecured interbank lending market

“…The dynamics of our estimated model confirm that shocks to counterparty-risk uncertainty can reduce lending activity for extended periods of time that are also accompanied by a more concentrated lending network. In this latter respect, Gabrieli and Georg (2014) provide empirical evidence on the network shrinkage in the euro money market during the 2007-2008 financial crisis.…”

Contagion in Stable Networks

“…4 For excellent literature reviews see Acemoglu et al 2017a, Babus and Allen (2009) and Glasserman and Young (2016). 5 For a network approach to market freezes see Gabrieli and Georg (2014). 6 See Shleifer and Vishny (2011) for a review of the literature.…”

“…It follows that relaxing this constraint should not alter our conclusion that incomplete but connected networks are not stable. 24 However, the real-life directed economic networks that have the general structure of our model (e.g. input-output and …nancial) are connected but incomplete.…”