Central Banking

Principles for financial plumbers

Payment systems, the "plumbing" underpinning international capital markets, aretoo often ignored. John Trundle from the Bank of England, who recently helpeddevise new standards for payment system safety, explains what changes areneeded in the latest issue of The Financial Regulator.
Payment systems, the "plumbing" underpinning international capital markets, aretoo often ignored. John Trundle from the Bank of England, who recently helpeddevise new standards for payment system safety, explains what changes areneeded in the latest issue of The Financial Regulator.

Email: John.Trundle@bankofengland.co.uk

_______________________________________________________

Payment systems are not the sexiesttopic in the world. I know well theglazed look that comes acrosspeople's faces when they have made themistake of asking me what I do for a living(we rapidly move on to anecdotes aboutbanknotes). But before you skip on to theenlightening and entertaining interview withHoward Davies, you might like to ponderon the world's financial plumbing for amoment. When conventional plumbinggoes wrong, there is nothing moreimportant than getting it fixed immediately,and then finding out why it was notdesigned or built properly in the first place.So it is with the plumbing of the world'sfinancial markets: payment systems.Central bankers around the world are tryingto ensure that at least the most significantsystems follow the best designs and are builtand operated reliably.

Central bankers on the case

In 1998, the Committee on Payment andSettlement Systems (CPSS) of the G10central banks, based in Basel, set up a taskforce to establish what good design andoperation of payment system meant inpractice. The task force comprised paymentsystem experts not only from G10 centralbanks, but also representatives from afurther eleven central banks (representingcountries in different stages of economicdevelopment), and from the InternationalMonetary Fund (IMF) and the World Bank.This exercise is one strand of the work tostrengthen international financialarchitecture to make it more robust and ableto withstand shocks such as those seenduring the international financial crisisbeginning in Asia in the autumn of 1997.

The CPSS-led group, which I chair,produced a consultative report in December1999 setting out ten core principles forsystemically important payment systems andfour responsibilities for central banks inimplementing those core principles. Theseare shown in the box (see p.46). A secondpart to the report was issued for consultationin June 2000 providing further detail andguidance on how to implement the coreprinciples. Following many detailedcomments on that report, the group iscompleting its work and hopes that afinalised report will be published at the beginning of 2001 (The consultative report is available from the Bank for International Settlements' website at www.bis.org).

The power of consensus

At the outset, the task force decided that itwanted to establish what was commonground internationally about the desirablecharacteristics of payment systems. Whatwas not then clear was whether a "one sizefits all" approach would work in practice.By confining our attention to significantpayment systems, which we defined assystemically important payment systems(SIPS), it did work. SIPS are paymentsystems which have the capacity, if not well-designedor well-run, to transmit shocks fromone participant to another. The reportdiscusses what makes a system systemicallyimportant. It may be because of the value oftransactions going through the system or thenature of the transaction (eg if it settles otherclearing systems such as those used insecurities markets). This approach enabledthe task force to establish principles whichwere both demanding and universallyapplicable to SIPS around the world. Onthat basis, it became clear that there was awidespread consensus on the principlesbetween central banks, commercial bankersand payment system operators, and acrosscountries.

The task force based its efforts on earlierwork by G10 payment experts and inparticular on the Lamfalussy standards,published in 1990. These had beendeveloped for a specific application -interbank netting schemes - but proved tobe appropriate for SIPS more generally withonly a little modification. The task force alsodecided that it was useful to elaborate onand add to some of the ideas contained inthe original work; there are ten coreprinciples as against the original sixLamfalussy standards.

In its early work the task force conducteda series of workshops with central banks indifferent parts of the world to develop thecore principles and test them as a descriptionof the international consensus. It was veryencouraging that they proved so robust. Thecore principles are the same throughout theworld, but the precise way in which theymight be implemented in countries withdifferent institutional, geographical andeconomic characteristics are different.

Safety vs. efficiency

Although the title of this article emphasizessafety, the public policy objectives forpayment systems are that they should be safeand efficient. In the developed worldefficiency is often taken largely for granted asbeing likely to be achieved by marketoutcomes.

Central banks in developed countriestypically seek to ensure that the safetyobjective is not under-emphasised forexample because the consequences maynot be borne fully by those who bring risk tothe system. In emerging markets, efficiencyis sometimes an equally pressing concern.There may be a need to develop a reliablepayment infrastructure to support economicdevelopment perhaps because only arudimentary structure exists at present. Ineither case it is necessary to take account ofany trade-off between the objectives inmaking proposals for reform. Clearly, if it ispossible to achieve more safety or moreefficiency without damaging the otherobjective, such change is desirable. But ifno such further gains can be made, choiceswill have to be made between objectives.Typically, central bankers will emphasize thesafety objective but not without regard tothe cost in terms of efficiency. In practice itis often possible to make such choices bylooking at comparators in similar countriesand adopting current best internationalpractice. Some specific examples arediscussed in the second part of the report.

The core principles

The ten core principles cover legal, financialand operational risk as well as efficiency.Five of the principles relate to financial risks.Six of the core principles (1, 2, 3, 5, 7 and9) are the same as, or very similar to, theLamfalussy Standards. Legal risk is listed firstbecause without a sound and predictablelegal basis, for example without explicitcontract law, it is impossible to managefinancial risks. The financial risk principlesrelate to the design of the system to ensurethat financial risk is contained, understoodand managed. The operational riskprinciple is perhaps most easily understood.The access and efficiency core principles arerelated and recognise the potential trade-offbetween the safety and efficiencyobjectives. The final core principle relatesto governance. Effective governanceunderpins all the other core principlesbecause governance determines the way inwhich decisions are made andimplemented.

RTGS systems

I am often asked why we did not have anexplicit principle that countries should havea real-time gross settlement (RTGS) system.Such systems are an excellent way to reducefinancial risks in general and to eliminatecredit risk in particular. They are tried andtested, having been introduced successfullyin more than 40 countries around the world.I personally have been a staunch advocate ofthem. But they are a means to the end ofachieving the core principles, not part of theend itself. They are particularly good atachieving prompt (in fact real-time)settlement finality - the fourth coreprinciple. But this principle can be satisfiedin other ways. The pace of innovation hasbeen accelerating; systems are beingdeveloped which use offsetting mechanismsto execute payments simultaneously,reducing the liquidity demands onparticipants. Whether this is a usefulalternative will depend, for example, on theexplicit or implicit costs of obtaining liquidityand on the robustness of the legalarrangements underpinning the offsettingprocess. Clearly all the other core principlesalso need to be taken into account inevaluating alternatives. The purpose of thecore principles was not to provide a universalblueprint for the design and operation ofpayment systems but to provide a frameworkagainst which they could be assessed.

Oversight responsibilities

In addition to the ten core principles the taskforce has published four responsibilities forcentral banks in implementing the coreprinciples. Throughout the world centralbanks have a central role in, andresponsibility for, payment systems. Thenature of those roles varies enormously. Inparticular, the role of overseeing paymentsystems varies from, at one end, countrieswhich have a formal statutory basis andexplicit powers for oversight (as is the case inAustralia and for the European Central Bank)to those whose approach is based on customand practice with few (or no) formal powers(as is the case in the UK and in Japan). Manycountries fall in between. Nonetheless, theapproach is typically less formal than thattaken towards the regulation of individualfinancial institutions such as banks.

Clear objectives needed

In recent years there has been a trendtowards greater formality in the oversightprocess, perhaps reflecting increasedconcern about systemic risks; many centralbanks have acquired new powers andresponsibilities (these include Belgium,Canada, Italy, Mexico and South Africa).The task force did not think it appropriate tosuggest any specific model for the oversightprocess. It did, however, conclude that eachcountry, within its own constitutionalstructure, should make clear both theobjectives of its oversight and the way that itcarries it out. The ECB, Belgium, Canada,Italy and the United States, among others,have all published statements of oversightpractice. [The Bank of England alsopublished a statement about its oversight ofpayment systems in November 2000.] TheBank of England's approach is based on co-operationwith the operators andparticipants in the main payment systems.

The Bank's principal focus is on systemic riskand therefore on systemically importantpayment systems. It has powers under theregulations implementing the SettlementFinality Directive to designate paymentsystems to provide greater certainty aboutthe applicability of the system's rules relatingto finality. The Bank has designated the twohigh-value payment systems in the UK fortransactions in sterling and euro (CHAPSsterling and CHAPS Euro). While the mainfocus of the Bank's oversight effort is onsystemic risk, the Bank also has an interest inpayment systems which may create "system-widerisk"; such systems do not involve largeinter-bank exposures but are used verywidely and have no practicable substitutemechanism, at least in the short term. Themain approach of the Bank's oversight roleto date has been in ensuring that thefinancial risks are contained through gooddesign but it is currently discussing with these systems their approach to containingoperational risks.

The second of the central bankresponsibilities relates to systems theyoperate themselves. It is very common forcentral banks to operate a country's mainhigh-value payment system and centralbanks also operate other systems such ascheque systems and automated clearinghouses. Where these are systemicallyimportant, the central bank should ensurethat they comply with the core principles. Inthe case of SIPS not operated by the centralbank, the central bank should overseecompliance with the core principles by thosesystems, and needs to have the ability tocarry out that oversight. The fourthresponsibility recognises thecomplementarity between the oversight roleof the central bank and activities by otherregulators, such as banking regulators orcompetition authorities, and encouragescentral banks to co-operate with othercentral banks and any other relevantdomestic or foreign authorities.

Practical implementation

After the publication for consultation of thefirst part of the task force's report, manyrespondents asked for further help inimplementing the core principles. Thesecond part of the report seeks to meet thatneed by giving examples of specific issuesthat can arise in seeking to satisfy the coreprinciples and how they have been met insome circumstances. In practice the processof implementation is likely to involve thepayment system operator (which could bethe central bank) the participants in thesystem, and the central bank as overseer.A good place to start is with a stock-takeof the principal payment systems in acountry. It is then necessary to assess whichone(s) are systemically important. The reportgives some guidance on relevant factors. It isvery likely that the system processing thehighest value of transactions will besystemically important but there may bemore than one SIPS. Once the systems havebeen identified the core principles can beused as a checklist to assess them.

A survey carried out in 1998 as part of aresearch project at the Bank of Englandsuggested that many payment systemsaround the world had little or no risk-proofing.The most common type of systemwas an unprotected deferred net settlementsystem which, without risk controls, canexpose participants to substantial financialrisks. In part, this result reflected the factthat, in many countries, the systems used forretail payments were often also used forhigher value wholesale payments, forexample where the only non-cash paymentsystem is a cheque system. Cheque systemsare a popular and effective mechanism formeeting many retail payment needs but, ifthe system is used frequently for high valuepayments, explicit risk controls will beneeded. In some cases, it may be easier toseparate the higher value payments andsettle them in a different paymentmechanism such as a credit transfer system.The report discusses some of the specificissues raised by cheque systems.

Legal issues

Another issue which commonly needs to beaddressed in both developed and emergingcountries arises from imperfections in, oruncertainty about, the adequacy of the legalstructure. Examples include countries wherethe legal structure may not adequatelysupport netting arrangements or paymentfinality. Operational risk issues may ariseeither because of the structure of thepayment system itself or because of itsreliance on other vulnerable nationalinfrastructure (such as telecommunicationsor electricity).

Where systemically important paymentsystems do not currently comply with thecore principles, a plan is needed to meetthem as soon as practicable. A central bankmay well have a leading role in such aprocess either as operator, settlement agentor as overseer. But any plan is likely toinvolve many parties and it will be importantto ensure that all parties are committed tothe process. Many countries have used anational forum as a mechanism to achievebuy-in to the process and to resolve practicalimplementation problems in the course ofthe project.

In some countries existing systems mayalready satisfy the core principles. But it isimportant to keep compliance under review.For example, with growing volumes, andwith greater use of complex technology,operational risk has become an increasinglyimportant concern. This issue was high-lightedby the assessment of the risks in thetransition to the Year 2000. Many paymentsystems found programming errors in theirsoftware which would have made systemsoperate incorrectly, or in extreme cases fail,in the Year 2000. It required an extensiveprogramme of work before the end of 1999to ensure that as many as possible of thesebugs had been identified before thetransition so that the operational teams coulddeal with any remaining bugs that slippedthrough the net. In practice this transitionwas managed very effectively throughout theworld but the process revealed manypractical insights into operational risks. Itbecame apparent that the contingency planswere often moderately well advanced for thephysical threats such as fire, flood or bomb atthe main processing sites. Such plans reliedon having identical processing sites with up-to-date data elsewhere. This approach,however, is not effective against a softwareproblem where the software problem iscopied in the back-up sites. Moreover, theprocess revealed that the contingency plansoften assumed that all other services wouldbe available. In the Y2K work it wasnecessary to develop flexible plans withdifferent options to enable operators tochoose from a menu of responses to anoperational problem depending on whatother difficulties were occurring at the sametime. Such planning can be extremelyfruitful when a real emergency happens.

Operational reliability is an importantaspect both of safety and efficiency.Operational problems are perfectly normal.They happen every day, especially incomplex systems. The key to ensuring thatthe risks are contained is that the operatorand participants have plans to cope withproblems and can implement businesscontinuity measures quickly and effectively.Work in this area is likely to becomeincreasingly important as the adoption ofgood design throughout the world reducesthe direct financial risks inherent in manysystems leaving legal and operational risks asthe focus of attention.

Next steps

It is important that no country should rest onits laurels. Not only can many aspects ofpayment systems be improved continuouslybut the systems themselves, and theenvironment in which they operate, aresubject to constant change. The possibilitiesfor promoting safety and improvingefficiency are always changing. For examplethe use of the internet and internet-basedprotocols are offering exciting possibilities forproviding efficient payment systems. Suchdevelopments raise questions about thesecurity and reliability of systems using thembut the concerns are being tackled - holdingout the prospect of safer and more efficientsystems in the near future. The technicalchoices available will change over time butthe principles to be satisfied will not. Ifpayment system operators and theiroverseers are successful in ensuring that theyare met, the study of financial plumbing canremain a subject for enthusiasts only.

John Trundle heads the market infrastucturedivision of the Bank of England. He is amember of the CPSS and of the equivalentECB committee which guides the TARGETsystem. John chairs the group established bythe G10 which produced the core principlesfor systemically important payment systems.

Email: John.Trundle@bankofengland.co.uk

Only users who have a paid subscription or are part of a corporate subscription are able to print or copy content.

To access these options, along with all other subscription benefits, please contact info@centralbanking.com or view our subscription options here: http://subscriptions.centralbanking.com/subscribe

You are currently unable to copy this content. Please contact info@centralbanking.com to find out more.

FedNow – at last

The instant payment system might help fix the US’s rusty payment rails, but it also faces competition, says Dave Birch

You need to sign in to use this feature. If you don’t have a Central Banking account, please register for a trial.

Sign in
You are currently on corporate access.

To use this feature you will need an individual account. If you have one already please sign in.

Sign in.

Alternatively you can request an individual account

.