Liquidity Networks, Interconnectedness, and Interbank Information Asymmetry
40 Pages Posted: 11 May 2020 Last revised: 2 Mar 2021
Date Written: February 28, 2021
Network analysis has demonstrated that interconnectedness among market participants results in spillovers, amplifies or absorbs shocks, and creates other nonlinear effects that ultimately affect market health. In this paper, we propose a new directed network construct, the liquidity network, to capture the urgency to trade by connecting the initiating party in a trade to the passive party. Alongside the conventional trading network connecting sellers to buyers, we show both network types complement each other: Particularly when information asymmetry in the market is high, liquidity networks reveal valuable information, provide a more comprehensive characterization of interconnectivity in the overnight-lending market, and improve short-term forecasts of soft information and country-specific yield spreads. We also compare the information content of both networks with that of traditional volatility and volume measures. We find that in normal market conditions when interconnectedness is high, a further increase in connectivity of either network raises volatility. In crisis periods when interconnectedness is low, however, an increase in liquidity network connectivity reduces volatility and boosts traded volume. Our results reveal the dual nature of interconnectedness—too much interconnectedness may increase systemic risk, but too little may impede market functioning.
Keywords: banking networks, interconnectedness, liquidity
JEL Classification: G21, C10, G10
Suggested Citation: Suggested Citation