Working Paper: CEPR ID: DP8859
Authors: Viral V. Acharya; Ouarda Merrouche
Abstract: We study the liquidity demand of large settlement (first-tier) banks in the UK and its effect on the Sterling Money Markets before and during the sub-prime crisis of 2007-08. Liquidity holdings of large settlement banks experienced on average a 30% increase in the period immediately following 9th August, 2007, the day when money markets froze, igniting the crisis. In the UK, unlike in the US until October 2008, the remuneration of reserves accounts provides strong incentives for banks to park liquidity at the central bank rather than lend in the market. We show that following this structural break, settlement bank liquidity had a precautionary nature in that it rose on calendar days with a large amount of payment activity and for banks with greater credit risk. We establish that the liquidity demand by settlement banks caused overnight inter-bank rates to rise and volumes to decline, an effect virtually absent in the pre-crisis period. This liquidity effect on inter-bank rates occurred in both unsecured borrowing as well as borrowing secured by UK government bonds. Further, using bilateral data we show that the effect was more strongly linked to lender risk than to borrower risk.
Keywords: cash; contagion; counterparty risk; funding risk; money markets; rollover risk; systemic risk
JEL Codes: E42; E58; G21; G28
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| liquidity demand of large settlement banks (E41) | overnight interbank rates (E43) |
| liquidity demand of large settlement banks (E41) | volumes in interbank market (G15) |
| precautionary liquidity hoarding (E44) | overnight interbank rates (E43) |
| precautionary liquidity hoarding (E44) | volumes in interbank market (G15) |
| liquidity effect during the crisis (E44) | borrowing costs for banks (G21) |
| higher liquidity demands (E41) | higher prices charged by lenders (G21) |
| greater funding and solvency risks (G32) | liquidity hoarding behavior (E41) |
| pre-crisis negative correlation (E32) | post-crisis positive correlation (G01) |