Share these talks and lectures with your colleagues
Invite colleaguesEstimating the contribution of liquidity sources in the Colombian large-value real-time gross settlement payment system: A preliminary approach
Abstract
This paper presents an empirical estimate of financial institutions’ use of the different liquidity sources for settling their obligations in the Colombian large-value payment system, called CUD. Based on that analysis, the paper also assesses the impact of various policies adopted by the central bank to foster earlier settlement of transactions and to mitigate liquidity risk in the real-time gross settlement system. To this end, CUD participants’ intraday transactional and liquidity behaviour was reviewed over two periods that differ considerably in both functionality of the system and economic environment: May 2002 and May 2010. The findings are presented in an aggregated manner by broad groups of participants, focusing attention on the three most active groups: banks, brokerage firms and trust companies. Owing to current limitations in available information, the methodology is based on the simplifying assumption that participants draw down the full balance on their individual reserve-requirement accounts at the central bank before turning to other liquidity sources. Next, it estimates the liquidity provided by the central bank or obtained in the money market and regards the payment flow between agents (liquidity from incoming payments) as a residual. This analysis provides a preliminary approach to identifying financial institutions’ strategies for using each of the alternative liquidity sources in the payment system and to determining the potential risks that such strategies pose to each type of participant and to the system as a whole. It also provides the financial authorities with criteria for designing macroprudential policies to mitigate liquidity risk and systemic risk. The authors believe this to be innovative research for emerging economies.
The full article is available to subscribers to the journal.