Variations in liquidity provision in real-time payment systems

We describe methods for measuring liquidity provision that can be applied to real-time gross settlement payment systems. Using data from CHAPS, the UK large-value payment system, we find that smaller banks tend to provide more liquidity than larger banks, relative to their payment flows. We use a Gini coefficient to measure these variations in liquidity provision between banks, and observe that the variations increase following the collapse of Lehman Brothers. It can be difficult to tell whether the variations are intentional or whether they occur due to external factors that are beyond the control of the individual banks. We use a recombinant approach to detect instances where observed patterns of liquidity provision are unlikely to have occurred absent of some behavioural or structural factors, such as differences in banks’ business models. Our results suggest that the variations in liquidity provision are larger than would be expected from truly random payment flows.