New Developments in Large-value Payment Systems

Transcription

1 New Developments in Large-value Payment Systems LUCHIA CHRISTOVA Bank for International Settlements Payment Conference of CIS Central Banks September 2005, St. Petersburg 1

2 CPSS established a working group aiming to analyse and describe the new developments since 1997 RTGS report, the core elements of the new settlement algorithms and of offshore payments systems. the report was published in May 2005 ( 2

3 Structure of the report 1. Elements of large-value payment system (LVPS) design 2. External influences on LVPS 3. Risk and costs in LVPS 4. Implications of new developments in LVPS 5. Possible future developments Annexes Comparative tables on selected LVPS The LVPS efficient frontier Framework for cost accounting 3

4 Country Acronym Year of implementation Annual number of transactions (2003; in thousands) Annual value (2003; in billions of USD) Belgium ELLIPS* ,760 15,306 Canada LVTS ,139 22,517 France TBF * , ,746 France PNS ,332 20,294 Germany RTGS plus* , ,115 Hong Kong HKD CHATS ,508 11,207 Hong Kong USD CHATS ,236 Hong Kong Euro CHATS Italy BI-REL * ,423 27,953 Japan BOJ-NET , ,914 Netherlands TOP * ,717 24,119 Singapore MEPS ,132 5,658 Sweden K-RIX ,302 13,900 Sweden E-RIX * ,141 Switzerland SIC ,700 33,202 United Kingdom CHAPS Sterling ,215 84,267 United Kingdom CHAPS Euro * ,292 35,227 United States Fedwire , ,706 United States CHIPS , ,561 European Union TARGET , ,993 European Union EURO ,852 50,501 European Union EPM * International Germany/Switzerland CLS SECB/euroSIC ,583 2, ,

5 New LVPS since 1997 European Union: TARGET, EURO 1 France: TBF, PNS Germany: RTGS plus Hong Kong: HKD CHATS, USD CHATS, EUR CHATS Singapore: MEPS Sweden: E-RIX United Kingdom: CHAPS Euro United States: NewCHIPS Switzerland: eurosic International: CLS 5

6 The life cycle of a payment Processing Conditionality, Queue release, Settlement LVPS Time Submission of payment order Irrevocability of payment order Finality of payment Transfer of the settlement asset 6

7 What has changed? Development of new systems Changes in the design and operations of the existing systems (BI-REL, newchips) Achieving intraday finality without an RTGS Introduction of offsetting algorithms in RTGS Reforms in intraday liquidity provisions by central banks (SIC, Scandinavian Cash Pool) Liquidity control measures More real-time information Increasing degree of interaction with other market infrastructures (DvP, PvP, intraday funds transfer between LVPS) New arrangements in correspondent banking Increased reliance on SWIFT 7

8 Driving forces for developments in LVPS Environmental factors: technological progress financial market liberalisation regional integration Users demand for enhanced services and risk control Encouraged by overseers of payment systems 8

9 Changes in payment systems design From pure RTGS unconditional payments based on available funds only transaction-oriented, each payment is processed one by one immediate settlement, provided that sufficient funds are available Advantages No credit risk Immediate finality in central bank money but High liquidity needs, thus high liquidity risk 9

10 and protected DNS conditional payments/ settlement lag continuous calculation of balances negative balances allowed within bilateral or multilateral limits typically end-of-day settlement of net balances (end-of-day finality) Advantages liquidity-saving through bilateral or multilateral offsetting but Settlement lag can result in credit, liquidity and systemic risk (participant banks implicitly grant each other credit during the day) Loss-sharing agreements (Lamfalussy protected system/survivors-pay) 10

11 DNS Lower bound Upper bound Delay Pure RTGS Liquidity available/required Unprotected DNS is cheaper because only net amounts are settled RTGS is safer because settlement is immediate 11

12 to LVPS with advanced features Two types of LVPS with advanced features : net settlement combined with a real-time functionality (intraday finality) safer than traditional DNS systems but probably more costly or RTGS systems with continuous offsetting as safe as traditional RTGS and probably less costly in terms of liquidity 12

13 DNS systems adopt RTGS features NewCHIPS, PNS RTGS systems adopt DNS features - RTGS plus, BI-REL DNS Lower bound Upper bound Delay Pure RTGS Liquidity available/required 13

14 As a result LVPS settle faster (intraday finality) with a lower amount of liquidity (offsetting), mainly central bank money at a lower liquidity cost (collateral policy) better control by participants over the settlement process 14

15 Disadvantages High development costs Less transparency? Is liquidity really a problem? 15

16 Conclusions No universally optimal LVPS design The report does not recommend any specific design 16

17 Cross-border use of collateral Cross-border collateral - collateral denominated in foreign currency or located in foreign jurisdiction Scandinavian Cash Pool Introduced in 2003 mostly for CLS payments Norwegian, Danish, and Swedish krona Use collateral located in one country to raise liquidity in another country Acceptance of foreign currency denominated collateral USA, UK, Sweden, Switzerland Correspondent Central Banking Model (CCBM) Introduced in 1999, participants of TARGET Post collateral at foreign CB to obtain credit from the home CB 17

18 New dimensions in correspondent banking Hong Kong USD and euro systems USD CHATS Euro CHATS Euro payment system for Swiss banks eurosic Advantages higher standardization (STP) settlement in real time with immediate finality possibility to achieve DVP and PVP 18

19 Risks and costs in LVPS Tradeoff between overall risks and overall costs efficient frontier Payment system risks - tradeoff between one type of risk vs another 19

20 The LVPS efficient frontier Overall risk A C D Upper limit of risk B Overall costs 20

21 Risks in LVPS Settlement risk can be influenced by LVPS design Settlement conditions Settlement delay 21

22 Risks and settlement delay Depends on participants behavior The earlier payments are settled, the earlier settlement risk is eliminated Incentive to delay payments in order to use incoming funds Could cause slowdown or even gridlock System designs to reduce settlement delay Throughput requirements Sender limits Time-dependent transaction fees Offsetting algorithms Non-binding behavioral conventions or implicit contracts 22

23 Costs in LVPS Costs for central banks Participants cost Operators cost 23

24 Possible future developments Continuing technological advancement Increase of processing power (LVPS and retail systems) Expansion of range of feasible business continuity arrangements Changes in financial market structure Increase of time-critical payments Demand for cross-border and multi-currency settlement services (TARGET2) Competition from large correspondent banks 24

25 Thank you for your attention! 25