Emergence of networks in large value payment systems (LVPSs)

This paper develops and simulates a model of emergence of networks in an interbank, RTGS payment system. A number of banks, faced with random streams of payment orders, choose whether to link directly to the payment system, or to use a correspondent bank. Settling payments directly via the system imposes liquidity costs, which depend on the maximum liquidity overdraft incurred during the day. On the other hand, using a correspondent entails paying a flat fee, charged by the correspondent to recoup liquidity costs and to extract a profit. We specify a protocol whereby banks sequentially choose whether to link directly to the system or to become clients of other banks, thus generating a client-correspondent network. We calibrate our model on real data on the UK payment system, and we compare the networks it produces with i) the true client-correspondent network, ii) the outcomes of two ‘dummy' benchmark models. The model is found to outperform the benchmarks. Its predicted networks reproduce some key features of the real UK network.