Identifying systemically important banks in payment systems


Published on

Talk at 'Systemic Risk Conference - Economists meet Neuroscientists' in Frankfurt on 18 September 2012. The conference was organized by House of Finance and Frankfurt Institute for Advanced Studies.

Published in: Economy & Finance, Business
  • Be the first to comment

No Downloads
Total views
On SlideShare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide

Identifying systemically important banks in payment systems

  1. 1. Systemic Risk ConferenceEconomists meet NeuroscientistsFrankfurt, 18 September 2012Identifying Systemically ImportantBanks in Payment SystemsKimmo Soramäki, Founder and CEO, FNA ( Cook, Chief Scientist, FNA
  2. 2. Interbank Payment Systems• Provide the backbone of all economic transactions• Banks settle claims arising from customers transfers, own securities/FX trades and liquidity management• Target 2 settled 633 trillion in 2010
  3. 3. Systemic Risk in Payment Systems• Credit risk has been virtually eliminated by system design (real-time gross settlement)• Liquidity risk remains – “Congestion” – “Liquidity Dislocation”• Trigger may be – Operational/IT event – Liquidity event – Solvency event• Time scale is intraday, spillovers possible
  4. 4. Agenda• Centrality in Networks• SinkRank• Experiment and Results• Implementation
  5. 5. Centrality in Networks
  6. 6. Common centrality metricsCentrality metrics aim to summarize some notion ofimportanceDegree: number of linksCloseness: distance from/to othernodes via shortest pathsBetweenness: number of shortestpaths going through the nodeEigenvector: nodes that are linked byother important nodes are more central,probability of a random process
  7. 7. Eigenvector Centrality• Problem: EVC can be (meaningfully) calculated only for “Giant Strongly Connected Component” (GSCC)• Solution: PageRank
  8. 8. PageRank• Solves the problem with a “Damping factor” which is used to modify the adjacency matrix (S) – Gi,j= Si,j• Effectively allowing the random process out of dead-ends (dangling nodes), but at the cost of introducing error• Effect of – Centrality of each node is 1/N – Eigenvector Centrality – Commonly is used
  9. 9. Which Measure for Payment Systems?
  10. 10. Centrality depends on process• Trajectory • Transmission – Geodecis paths (shortest paths) – Parallel duplication – Any path (visit no node twice) – Serial duplication – Trails (visit no link twice) – Transfer – Walks (free movement) Source: Borgatti (2004)
  11. 11. Distance to Sink• Markov chains are well-suited to model transfers along walks• Absorbing Markov Chains give distances: From B 1 To A From C 2 From A To B From C 1 From A To C From B
  12. 12. SinkRank SinkRanks on unweighed• SinkRank is the average distance networks to a node via (weighted) walks from other nodes• We need an assumption on the distribution of liquidity in the network at time of failure – Asssume uniform -> unweighted average – Estimate distribution -> PageRank - weighted average – Use real distribution -> Real distribution are used as weights
  13. 13. SinkRank – effect of weights Uniform PageRank “Real” (A,B,C: 33.3% ) (A: 37.5% B: 37.5% C:25%) (A: 5% B: 90% C:5%) Note: Node sizes scale with 1/SinkRank
  14. 14. How good is it?
  15. 15. Experiments• Design issues – Real vs artificial networks? – Real vs simulated failures? – How to measure disruption?• Approach taken 1. Create artificial data with close resemblance to the US Fedwire system (BA-type, Soramäki et al 2007) 2. Simulate failure of a bank: the bank can only receive but not send any payments for the whole day 3. Measure “Liquidity Dislocation” and “Congestion” by non-failing banks 4. Correlate 3. (the “Disruption”) with SinkRank of the failing bank
  16. 16. Distance from Sink vs Disruption Relationship between Failure Distance and Disruption when the most central bank fails Highest disruption to banks whose liquidity is absorbed first (low Distance to Sink) Distance to Sink
  17. 17. SinkRank vs Disruption Relationship between SinkRank and Disruption Highest disruption by banks who absorb liquidity quickly from the system (low SinkRank)
  18. 18. Implementing SinkRank
  19. 19. Implementation example available at
  20. 20. More information Soramäki, D.Sc.kimmo@soramaki.netTwitter: soramakiDiscussion Paper, No. 2012-43 | September 3, 2012 |
  21. 21. Data generation processBased on extending Barabasi–Albert model of growth andpreferential attachment