Intraday liquidity needs in a modern interbank payment system : A simulation approach

Abstract

In this study three topics in particular are analysed.First, the adequacy of intraday credit limits in the existing and planned Finnish interbank funds transfers systems are studied.Secondly, the efficiency of two actual and two hypothetical payment settlement systems is analysed and, thirdly, the effects of optimization features such as queuing, payment splitting and netting of queued transfers are examined. The data used in the study consist of four days of actual payment data provided by the major banks operating in Finland and 100 days of generated data, where the order and number of daily payments is varied.The study is based on results generated by a payment systems simulator that was developed at the Bank of Finland.The simulator is capable of simulating a wide variety of hypothetical settlement and banking structures as well as various optimization features. The results show that the existing intraday credit limits of the Finnish banks will be sufficient in the settlement system to be used in Stage Three of Economic and Monetary Union.However, some queuing of payments is expected to take place if liquidity is not increased.Systems employing more real-time gross settlement were found to be more efficient in terms of speed of settlement and usage of liquidity per value of payments settled.An RTGS system operating with the same liquidity as an end-of-day net settlement system results in substantially faster settlement of payments. Optimization features enhanced the operation of the system significantly.Splitting of payments enhanced the circulation of liquidity in the system and prevented the formation of gridlocks, thus reducing settlement delay substantially.The main effect of the netting of queued transfers was that it solved gridlock situations during the day and thereby reduced settlement delay. Keywords: payment systems, clearing/settlement, liquidity, simulation, optimization methods

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 36 citations exist upstream.