The Impact of Operational Events on the Network Structure of the LVTS

Bank of Canada discussion papers are completed research studies on a wide variety of technical subjects relevant to central bank policy. The views expressed in this paper are those of the authors. No responsibility for them should be attributed to the Bank of Canada. useful comments and suggestions, and Geoffrey Hill for research assistance. A special thank you goes to Christian Belisle for his knowledge of operational events, and related participant behaviour, in the LVTS. Additional thanks goes to Mike Hoganson of the Canadian Payments Association. Abstract The author uses a quantitative network analysis approach to assess how participants in the Large Value Transfer System (LVTS) respond to partial outages at other banks. Despite the limited number of operational events, benchmarks can be established. For example, the effect of a partial outage at a big six bank that has a net payment balance of $1 billion is estimated to correspond to a decline in connectivity of about 6 or 7 per cent, due to other participants not sending payments to the problem participant. This suggests that participants tend to perceive a partial outage at a counterparty as inconvenient enough to warrant a delaying of payments, at least to the problem participant. There is no strong evidence of systemic effects, whereby participants also delay payments to non-problem counterparties, in an effort to maintain a reasonable liquidity position. Notable events mostly occurred in 2004 or 2005, and the incidence of operational events did not increase over the sample period. The improvement of contingency measures, such as the use of the LVTS Direct Network, possibly contributed in this regard. Résumé L'auteur utilise une approche quantitative d'analyse des réseaux afin d'évaluer comment les institutions qui font partie du Système de transfert de paiements de grande valeur (STPGV) réagissent en cas de panne partielle chez d'autres participants. Malgré le nombre limité d'incidents opérationnels à s'être produits, il est possible d'établir des balises. À titre d'exemple, on estime que l'effet d'une panne partielle dans une des six grandes banques canadiennes qui aurait un solde net de paiement de un milliard de dollars correspond à une perte de connectivité d'environ 6 ou 7 %, attribuable à l'interruption des paiements envoyés à cette institution par les autres. Un tel résultat indique que ce type d'incident est généralement assez préoccupant aux yeux des participants pour justifier un report des paiements, du moins à la contrepartie en cause. Rien ne …