The third edition of Bank of Finland series on payment and settlement system simulation studies has been published today. The preliminary versions of the papers have been presented at the annual simulator seminars arranged by the Bank in 2007 and 2008. The main focus of the analyses in this edition is on business continuity arrangements, operational stability, liquidity requirements and saving mechanisms, gridlock resolution, transaction queuing arrangements, and the topology of payment networks. The studies examine systems in several countries and cover different kinds of payment systems and regimes.

There are two papers co-authored by me and Marco Galbiati from the Bank of England in this volume. The first paper “An agent-based model of payment systems” looks at banks decisions to commit liquidity for settlement under different conditions. We find that banks tend to commit less liquidity than what would be optimal. The second paper “Liquidity saving mechanisms and bank behavior” extends the above model to an environment where banks can choose how much liquidity to use, and whether to submit payments for immediate settlement or a liquidity saving mechanism. We find that the potential benefits of liquidity saving mechanisms are not fully captured due to behavioral changes by the banks.

The book also contains a paper by Peter Zimmerman, Anne Wetherilt and me on the network topology of the sterling unsecured loan market during 2006–2008. The paper identifies overnight loans from payment data by matching the two legs of the transaction on subsequent dates (update: you can now download the tool). This gives us a unique data set which we use to see the impact of the crisis on this unsecured segment of the market. We find that the core banks have become more important during the crisis, and that the widened reserve target bands have allowed banks to exercise more discretion in forming lending and borrowing relationships.