The policy implications of electronic payments

Release date
16/05/1996
Speaker
Peter J. Ledingham

Presentation to the "Consumer Payment Systems" Conference

Introduction

The following paper was presented at a conference on "Consumer Payment Systems" in Auckland on the 16 May 1996. Some parts of the paper are drawn from an earlier article, "Pre-paid cards", Reserve Bank of New Zealand Bulletin December 1994

1. There have been a number of indications recently that the use of electronic methods for making "retail" payments may be about to expand significantly in scope. A number of stored value card projects are under way - some under discussion, and others being trialled or implemented in a number of different countries. There is growing interest in developing reliable methods of payment for use on the Internet. I will not attempt to survey these developments in any detail, as other speakers at this conference know far more about them than I do.

2. This paper focuses on some of the general issues which arise from these developments, rather than commenting on any particular proposal. It outlines the similarities and differences between emerging electronic payment methods and the current "conventional" payment instruments such as currency and debit cards. It goes on to examine some of the policy issues involved, such as the implications for the Reserve Bank's currency issue and prudential issues.

The characteristics of electronic payments

3. There are two basic methods of making payments, and the distinction between them is fundamental to much of the discussion in this paper. The first (the "account transfer" system) involves customers issuing an instruction to "banks" to debit the account of the person making payment, and credit the account of the person receiving it. Depending on the technology being used, these payments may sometimes be effectively "final" (from the customer's perspective) at the point when they are made (e.g. EFTPOS transactions). Other types of payments (notably cheques) are conditional, and may not become "final" for some days. Payment methods which fall into the account transfer category include cheques, debit cards, credit cards, and telephone banking.

4. The second generic method is the "direct transfer" or "token" system. Under this model, a form of "money" is established, which can be directly handed from one person to another, without the direct involvement of any "bank" in the transaction. These forms of money are purchased from an "issuer", who ultimately carries the obligation to redeem them. The transaction is "final" at the point when it occurs, and is not dependant on the operation of some underlying settlement system. Cash is the most widespread example of a token system operating, but in fact there are a number of examples. Some of them are "monetary" in nature - they are denominated in dollars (or some other currency). Cash, phone cards and gift tokens are common examples. Others are effectively "real" tokens - when used, they provide a given quantity of some good or service. The familiar 10-trip train and bus tickets are obvious examples of real token systems - and, indeed, their predecessors were perhaps the first "stored value cards" which gained widespread use.

5. It appears that pending "electronic purse" developments will fall into both of the above types. Some of them will be implemented using "smart" cards, and others on various networks. Some will involve "reloadable" purses, while others will remain disposable. It seems likely that increasing emphasis will be given to inter-operability: note that many of the existing token systems, in particular, are currently restricted to dealings between purchasers and the "issuer" (i.e. the issuer is always the recipient of any payment made).

6. Figure 1 sets out some of the technical characteristics of different payment methods. It includes a number of familiar examples, and also includes two generic examples of electronic payment developments - one of the account transfer type, and one of the direct transfer type.

Are electronic payments "money"?

7. Somewhat surprisingly, there is no single definition of what constitutes "money". Conventional Economics text books describe some of the different roles that money plays - typically as a medium of exchange, a unit of account and a store of value. In this paper, the focus is primarily on the "medium of exchange" role, although this cannot be entirely separated from the other roles. In passing, note that some of the technical characteristics which a good "medium of exchange" should have are divisibility, portability, and durability.

8. In thinking about whether electronic payments are "money", it is worth reflecting on the main attributes of currency (i.e. the notes and coin issued by the Reserve Bank):

(a) It is a standard product (or a set of standard products) and is therefore easily recognised and identified.

(b) It is risk-free, in the sense that the Reserve Bank stands fully behind it. (Of course, currency can be risky to hold in that it can be lost or stolen).

(c) It is fully negotiable. Possession of currency is sufficient to establish a right of use, and it can be simply handed to someone else in order to make a payment.

(d) Usage is anonymous and not directly traceable.

(e) It is convenient and efficient for making small payments.

(f) It is valid consideration in all places and circumstances, and (except where forgery is suspected) its status cannot be questioned (both legal tender rules and the law more generally imply that, when it comes to the crunch, this is the only type of "money" which really counts).

(g) Payments made using currency are certain. They are final and irrevocable at the point when they are made, and are not dependant on the operation of any clearing/settlement arrangements.

9. These attributes collectively imply that currency is generally accepted. However, it is not always the preferred method of payment, particularly for higher value transactions - since use of currency involves handling, storage and security costs which may not arise to the same extent with other methods.

10. Figure 2 sets out how some of the other payment methods compare in respect of these attributes. Clearly, some score well on some items and poorly on others. However, it is also the case that the relative importance of these various attributes is very dependant on the type of payment being made - where, how large and how often, for example. Cash can be very convenient in some situations, but a real nuisance in others - and the same is true for other technologies. Probably the key conclusion to draw is that different payment technologies are not inherently inferior/superior to one another. Rather, each of them has an appropriate place - as, indeed, the current happy co-existence of a number of overlapping/competing alternatives clearly indicates.

11. It is interesting to speculate on the potential impact of stored value cards and other new forms of payment on the existing technologies. Clearly, some substitution will take place, but the nature and extent of this substitution will depend on a number of factors. People will tend to prefer to use payment technologies which are cheaper, more convenient, and/or less-risky than available alternatives. Many will probably prefer methods which can be used for multiple purposes, rather than having to utilise a variety of methods (or cards) to meet different needs. The level of acceptance of particular payments by retailers, merchants and other suppliers - not just in New Zealand, but around the world - will obviously have an important influence on the take-up of new approaches. Exactly how such influences will work themselves out remains to be seen.

The policy issues

12. The Reserve Bank's main interests in new forms of electronic payments lie in two areas: the effect that they might have on the Reserve Bank's currency issue, and the possible implications for the safety and soundness of the financial system. These issues are discussed in turn, along with one or two other matters of interest.

Reserve Bank currency issue

13. The Reserve Bank Act gives the Bank the sole right to issue bank notes and coins in New Zealand. For banknotes, this has been the case since the foundation of the Reserve Bank, and the 1989 Reserve Bank Act extended the monopoly to include the coin issue as well (the coin issue had previously been a Treasury monopoly).

14. The conceptual case for public provision of currency is probably reasonably strong. The various attributes of currency cited above embody the reasons: there is a public benefit in having available a standard, reliable and efficient method for making certain (particularly small) payments.

15. However, the case for a statutory monopoly rests more on the seigniorage argument. Traditionally, at least, currency issue has been very profitable, and monarchs and governments have been keen to keep such a useful source of revenue to themselves. The profits have come from a combination of two features:

(a) The ability to issue currency which has a much higher face value than the actual production cost.

(b) The ability to invest at interest the proceeds of the currency issue - on which, of course, no interest is paid. (An alternative way to look at this is that currency issue can be a particularly cheap form of government borrowing)

16. Neither of these sources of profit is guaranteed in all circumstances. In fact, it is conceivable in a low-inflation low-interest-rate environment that the profit on the currency issue may be insufficient to cover the production and administrative costs involved.

17. In practice, the monetary authorities have been quite relaxed about seeing most of the "money" in the economy taking the form of commercial bank liabilities rather than central bank liabilities; and we have been relaxed about the banks (and others) developing payment methods (e.g. cheques, travellers cheques, debit cards and credit cards) which are close substitutes for the use of currency. Stored value cards and the like are simply another innovation in this context, and do not raise any new issues of principle.

18. Thus, Reserve Bank provision of currency under a statutory monopoly is seen as a useful public service, and one which provides a basis on which other payment arrangements and contracts must ultimately rest. But it is not thought to be an area where the Reserve Bank should seek to maximise its income. Rather, the Bank's role is to facilitate and encourage overall payment system efficiency by continuing to offer currency as just one payment technology amongst several. Alternative payment technologies and innovations can be freely allowed within this framework, and users can be allowed to choose freely amongst those competing technologies.

19. At this stage, there is no intention that the Reserve Bank should itself issue stored value cards, or provide other forms of electronic money to the public. However, the possibility that this might happen at some point in the future cannot be completely ruled out. If electronic developments proceed to the point where payment using currency becomes inefficient and costly, then it could prove to be necessary in effect to convert our paper and metal currency issue into electronic form in order to provide the public benefits that will continue to be necessary. One or two other countries have already decided to have some type of "government" issued stored value card, although perhaps not for this reason.

20. Finally in this section, it should be noted that electronic payments developments (and any erosion of the currency issue that may result) do not seem to pose any threat to the Reserve Bank's ability to operate monetary policy in the foreseeable future. Even if they did pose such a threat, the appropriate response would probably not be to seek to control or prevent such developments. It is generally much better to adapt the monetary policy operational framework to the realities of the marketplace than to seek to do the opposite.

Prudential issues

21. Earlier, it was indicated that there are some quite important distinctions between "currency" and privately issued "money". In particular, currency liabilities reside on the books of the Reserve Bank, and effectively carry a government guarantee. It is not known exactly where the liability for some of the pending payment technologies will rest, but they will obviously not carry any government guarantee.

22. It is important that members of the public fully appreciate this distinction. If they are told that something is an "alternative to cash" they may believe this, and fail to realise that the monetary amount stored on a card or on a network is only as good as the bank or other organisation which has the ultimate liability to pay on that obligation. Issuers will therefore need to be careful to represent their products as akin to a "deposit" rather than "cash". This concern may not be very serious with low-value balances, where the amount at risk for any individual is not very material. However, the concern will become more significant as higher-value systems emerge, and as options for re-loading of cards and/or third-party transfers of liabilities are introduced.

23. One of the reasons for this heightened concern is that it seems possible that issuers of electronic "money" could be in a risky business, because of the potential for counterfeiting. Depending on the technical characteristics of each system, counterfeiting could take place by physical reproduction of cards, by manufacture of a "re-loading mechanism" which could be fraudulently used to add new balances to authentic cards, or by "hacking" into the payments systems themselves. With some technologies, it might be difficult or impossible for the recipient of an electronic payment, or even for an issuer, to detect an electronic counterfeit. By contrast, counterfeit currency can often be detected by a recipient, and is always detected by the Reserve Bank - which never pays out on it. Sophisticated security features will mean that the risk of successful electronic counterfeiting is likely to be very low. However, given that the perceived returns could be quite high, the possibility cannot be ignored. The costs if successful counterfeiting occurred - for the issuer, and perhaps for some of its customers - could be devastating.

24. It is clear that issuers will need to take great care to ensure that the danger of counterfeiting is minimised, and that they are vigilant in monitoring their systems and operations so that it is quickly detected when it occurs.

25. In addition, it may be desirable that substantial electronic "issuers" (or issuers offering the ability to store "high" monetary values) should be subject to a disclosure regime, so that holders of their obligations can be made more aware of the financial condition of the issuers, and the potential risks they may face. It is likely that many actual or potential card issuers will be subject to disclosure requirements under the Securities Act or the Reserve Bank Act in any case, but there may be a need to fill any significant gaps in coverage which emerge.

26. Matters become a little more complicated when (say) New Zealand residents get involved in payment arrangements which are effectively domiciled outside New Zealand - presumably, some "Internet banking" activities could easily fall into this category. People will need to clearly understand who it is that they are dealing with, where they are "legally" located, whether they have appropriate "standing", and what potential risks they may face - either by holding balances within a system, or by using those balances in a transaction. This information may not always be easy to gather or evaluate.

27. Figure 3 summarises the basic risk characteristics of the different payment methods. Both the holders of new forms of electronic money, and the "issuers", face potential risks in certain circumstances. Some of these are reasonably easy to understand and evaluate, and may not be any more serious than the risks - such as losing your wallet - which exist at present. (Indeed, new technologies may help to reduce some of the risks that people bear at present.) However, the risk characteristics of schemes involving elements such as multiple issuers, many "bank" members, many "accepting" merchants, multiple jurisdictions, delay or complexity in any settlement arrangements, or lack of any audit trails may not be at all straightforward. These possible complexities are not dealt with in the table.

Standardisation and competition

28. The indications that a number of new payment technologies are likely to emerge raises some questions about the need for standardisation. The original EFTPOS developments demonstrated that competing alternatives can sometimes delay progress because retailers - and others - do not want to have to deal with a number of incompatible systems. At this point, it is not entirely clear whether the systems that are on the way will meet common standards, or whether developments will be more fragmented. However, this is primarily an issue for the various promoters to sort out - they have strong incentives to get it right themselves - and clearly a lot of effort has already been put in to this area. There is no obvious role for the Reserve Bank - or other Government agencies - in choosing or setting technical standards, or expressing any particular technological preferences. Indeed, any attempt to agree upon, or impose, common standards could work to prevent the best choices being made, and to deter subsequent innovation. Such costs would need to be weighed against the potential benefits involved.

29. A more general point here is that it is necessary to strike the right balance between co-operation and competition. Co-operation amongst different issuers and merchants may be necessary in order to achieve the degree of interchangeability which will attract people to use the new technologies. However, customer needs are likely to be best met through effective competition, and this will require that genuine choices amongst alternative products are available. The two goals are not necessarily incompatible, but they will not necessarily be achieved automatically or easily.

30. Figure 4 summarises some of the pros and cons of "standardisation". Perhaps the best overall conclusion to draw is that there are some solid reasons for thinking that standardisation is worthwhile; but that it needs to be kept in its place.

Statistics

31. Part of the Reserve Bank's role is to ensure that adequate information is available on the structure and operation of the financial system. Statistical and other information is an essential input into the monetary policy process, and into evaluation of the efficiency and soundness of the financial system. The Bank has long collected monetary statistics of various types, and it is important that these be kept up-to-date, and relevant, in the light of financial innovations of various types. Recent and pending payments system developments are posing some challenges in this respect. The Bank will be actively working with the various parties involved to ensure that its statistical collection and reporting arrangements are appropriately extended and adapted to meet the future information needs of both policymakers and market participants.

Legal matters

32. Any arrangements for making monetary payments require satisfactory legal underpinnings. It is important that all of the parties to a payment - the payer, the recipient, and any other parties such as banks or clearing systems who may be involved as intermediaries - clearly understand their rights and obligations, and any risks they may face. The rules need to specify clearly who carries risks at the various stages that a transaction may pass through, and what happens when things go wrong.

33. In practice, the overall rules which apply in a payment arrangement generally reflect a combination of statute law, common law, contractual arrangements and industry codes of practice. In the case of new electronic payments arrangements, it is likely that all of these will have some role to play. It may be possible to use current legal bases to create much, or even all, of the legal underpinning that is required. However, the distinctive characteristics of the new technologies may require some new thinking, and some new rules. In particular, where payments arrangements have significant cross-border elements (as some of them will), it may become very difficult to establish clearly the rights and obligations of the various parties that are involved in a transaction, the forms of redress available when things go wrong, and the jurisdiction which applies. In some cases, it may be a considerable challenge for promoters to be able to satisfy their customers that their systems are reliable, and that procedures which are straightforward and fair are available for the resolution of technical difficulties or any conflicts. The Bank will be monitoring developments in this area with considerable interest.

Conclusions

34. The Reserve Bank does not have any substantial concerns about the development of new retail payment methods at this stage. More widespread uses of these new methods will be financial innovations which are not radical from a conceptual viewpoint, and they do not raise any new issues of principle. However, they may raise a number of significant issues of practice, including risk management and legal issues. The Bank will be watching developments, with a view to ensuring that the integrity of the financial system is maintained.

Figure 1
Different payment methods - technical characteristics

METHOD

TYPE OF MONEY

ACCESS

FUNCTION

CUSTOMER SIGNATURE

ISSUER SIGNATURE

OPERATING COSTS

CASH

MONETARY TOKENS

PHYSICAL HANDOVER

DIRECT TRANSFER

NO

NO

HIGH

CHEQUES

BANK DEPOSITS

PHYSICAL HANDOVER

ACCOUNT TRANSFER

YES

YES

HIGH

DEBIT CARDS

BANK DEPOSITS

ATM
EFTPOS

ACCOUNT TRANSFER

PIN

YES

MEDIUM

BUS TICKETS

"REAL" TOKENS

PHYSICAL "CLIP"

DIRECT TRANSFER

NO

(YES)

LOW

PHONE CARDS

MONETARY TOKENS

PAYPHONES

DIRECT TRANSFER

NO

(YES)

LOW

SVC 1
INTERNET 1

MONETARY TOKENS

?

ACCOUNT TRANSFER

NO?

YES?

MEDIUM

SVC 2
INTERNET 2

MONETARY TOKENS

?

DIRECT TRANSFER

NO?

NO

LOW

Figure 2
Different payment methods - Are they "money"?

METHOD

STANDARD

MULTI-PURPOSE

NEGOTIABLE

PRIVACY

CONVENIENT EFFICIENT

LEGAL TENDER

CERTAIN

CASH

YES

YES

YES

YES

(YES)

YES

YES

CHEQUES

YES

YES

YES/NO

NO

(YES)

NO

NO

DEBIT CARDS

YES

YES

NO

NO

(YES)

NO

YES?

BUS TICKETS

YES?

NO

YES-WHOLE CARD

YES

(YES)

NO

YES

PHONE CARDS

YES

NO

YES-WHOLE
CARD

YES

(YES)

NO

YES

SVC 1
INTERNET 1

?

?

NO

NO

(YES)

NO

YES?

SVC 2 INTERNET 2

?

?

YES

YES

(YES)

NO

YES?

Figure 3
Different payment methods - Risk Characteristics

METHOD

GOVERNMENT GUARANTEED

COUNTERFEIT RISK TO ISSUER

COUNTERFEIT RISK TO HOLDER

HOLDER VULNERABLE TO LOSS/DAMAGE

CASH

YES

NO

YES

YES

CHEQUES

NO

NO

YES

YES

DEBIT CARDS

NO

NO

NO

NO

BUS TICKETS

NO

YES

NO?

YES

PHONE CARDS

NO

YES

NO?

YES

SVC 1
INTERNET 1

NO

NO

NO

NO?

SVC 2
INTERNET 2

NO

YES

NO

YES

Figure 4
Some Pros and Cons of standardisation

PROS

CONS

TECHNICALLY EFFICIENT

LOWEST COMMON DENOMINATOR

IMPROVED UNDERSTANDING

DIFFICULT TO CHANGE

COMMON UNDERSTANDING

HARD TO INNOVATE

LOWER RISK OF ERRORS

CONFLICTS OVER ACCESS

LOWER COST

CONFLICTS OVER CONTROL