Abstract

Co-authors: Gerardo Ferrara (University of Turin), Sam Langfield (ECB), Zijun Liu (Bank of England)

This paper studies systemic liquidity risk in UK interbank system in a dynamic context. We estimate the daily network structures of banks short-term funding (up to 30 days) with various confidential datasets, and identify (1) banks that fall short of liquidity by themselves (individually illiquid banks); (2) banks that fall short of liquidity because the counterparties fail to repay their debts to the banks (systemically illiquid banks); and (3) the timing of these banks falling short of liquidity. In order to consider the timing of defaults, not only the number of defaults which normal contagion models focus on, we test a dynamic financial contagion model for the first time in the literature. This dynamic feature is important particularly when we discuss liquidity regulations such as LCR .

We obtain outcomes by the numerical simulations of the dynamic contagion model, with assumptions consistent with PRAs micro-prudential liquidity monitoring schemes. The outcomes so far show the existence of significant systemic liquidity risk when banks do not have sufficient liquidity buffers. To the authors knowledge, this is the first paper in the literature studying interbank systemic liquidity risk with real datasets.

Based on the estimated systemic liquidity risk, we will consider the optimal liquidity provision policies to minimise the systemic liquidity risk, by solving a dynamic programming model.