It is a great pleasure for me to be able to speak to you about the legal questions arising out of the development of electronic funds transfer and, in particular, to tell you some about the work being carried out in this field by UNCITRAL, the acronym for the United Nations Commission on International Trade Law. The timing of this seminar could not have been more appropriate from our point of view, because in two months a working group of UNCITRAL will meet to examine a first draft of Model Rules for Electronic Funds Transfers that has been prepared by our secretariat. I must emphasize that this draft is the product of the secretariat; it has not as yet had any review by any committee or other intergovernmental body of UNCITRAL.* Nevertheless, its very existence and the fact that the process of considering it will begin in July are indications of the importance the subject has taken on and the response of UNCITRAL to some of the problems that are raised by this new phenomenon.

It is a great pleasure for me to be able to speak to you about the legal questions arising out of the development of electronic funds transfer and, in particular, to tell you some about the work being carried out in this field by UNCITRAL, the acronym for the United Nations Commission on International Trade Law. The timing of this seminar could not have been more appropriate from our point of view, because in two months a working group of UNCITRAL will meet to examine a first draft of Model Rules for Electronic Funds Transfers that has been prepared by our secretariat. I must emphasize that this draft is the product of the secretariat; it has not as yet had any review by any committee or other intergovernmental body of UNCITRAL.* Nevertheless, its very existence and the fact that the process of considering it will begin in July are indications of the importance the subject has taken on and the response of UNCITRAL to some of the problems that are raised by this new phenomenon.

The increased use of electronic techniques to make funds transfers through the banking system is of great importance to all concerned. That statement is easy to make, but what does it mean? How are electronic funds transfers (EFTs) important and who is concerned?

First, let us consider how they are important. EFT permits the handling of large volumes of funds transfers at lower costs than similar quantities of paper-based funds transfers. In some of its manifestations, EFT permits a faster processing of a funds transfer from the giving of the order by a paying customer to the crediting of the account of a receiving customer. It is now technically possible for even international funds transfers to be completed in fractions of a second. But this potentiality for speed raises questions about volatility of funds, and therefore about the effectiveness of monetary controls; it puts pressure on other elements of the funds transfer system, such as settlement systems, that may not currently be organized to permit maximum exploitation of the potentialities of EFT; it also permits—and therefore leads—banks and bank customers to rely on this faster speed (with which funds transfers can be made) in planning their own affairs. Commitments are made for the delivery of funds at a certain time and place in the expectation that the funds transfer process need be commenced only shortly before that time. It would defeat the very purpose of creating on-line systems if customers had to make their funds transfers a day or two early to avoid the consequences of errors by banks, as at least one important U.S. court has said they should.

This leads to the question of reliability. Fundamentally, the amount of time it will take for EFT to deliver funds to a certain account is considerably more predictable than is the amount of time it will take for paper-based techniques to do the same. This is a great advantage to all concerned. However, the reverse side of the situation is that, again, the transferors of fund come to rely on the reliability of the system. More is now at stake than there used to be if there is an unexpected delay or if there is an error. But EFT, in most of its configurations, is less robust than paper. With paper, if something goes wrong, there is usually some other way to carry out the transaction within the permissible period of time. With EFT, there is often less margin for error. If things go wrong, the potentiality for loss—even for catastrophe—is much greater.

All of these factors—and others—are important to you as central bankers. They are obviously important to the commercial banks in your countries. They are also important to the customers of banks. Those customers have a vital concern that the banking system be able to meet their needs to transfer funds—whether domestically or internationally—safely, securely, and according to understood and appropriate rules.

It is this last item that is of interest to UNCITRAL: whether electronic funds transfers, and especially international funds transfers, are made according to understood and appropriate rules. The evidence is that, to a distressing degree, they are not.

To explain the reason for this statement, let me describe briefly the situation in regard to the legal regime governing funds transfers. Until recently, countries could be divided into two major groups according to the means by which funds transfers were made. One group, apparently including all those with a legal and banking heritage from England, made funds transfers almost exclusively by use of checks. Funds transfers by way of telegrams or telexes between banks were certainly known, but were relatively rare. The law of checks was well established and in statutory form, since it either constituted part of the law of bills of exchange, as in the United Kingdom, the United States, and other common law countries, or was assimilated to it, as in all those countries that relied on the Geneva Uniform Laws of 1930 and 1931—uniform laws developed under the aegis of the League of Nations. There was, however, essentially no law governing telegraphic and telex transfers, a matter of little concern since they were relatively unimportant and few in number.

The other major group of countries either made all or essentially all funds transfers by giro or ordre de virement, as was the case in Germanic countries, or had a well-established system of giro or virement in addition to a check-clearance system, as was the case in France. The law governing giro or virement was well established in those countries, but it was not primarily in statutory form. Furthermore, and this is of particular importance today, there had never been any international effort to unify or even to harmonize the laws governing such transactions. Naturally, there were many elements in common in all legal systems. Common problems bring common solutions. Moreover, there had always been some interaction between banking systems and between experts in banking law, leading to the elimination of extreme differences in banking practice and law governing such funds transfers, but the generalization that the law governing giro or virement is national and non-uniform in nature is still valid.

This difference in history has had an effect on the perception of the nature of electronic funds transfers and, therefore, on the law governing them. In countries that have traditionally relied on giro transfers, the use of electronics has often been an incremental matter. That is, the customer who was transferring funds may have continued to hand over to his bank a payment order directing debit to his account and credit to the account of his doctor, his supplier, or his employee. If the account to be credited is at a different bank, the bank of the originator of the order might now keep the paper-based payment order and send forward the relevant data on a magnetic tape instead of (as it formerly did) sending the paper itself forward to the bank of the beneficiary. The customers of either bank have no reason even to know, much less to care, that the banks have changed their procedures. This gives the appearance that there is little reason to change the legal rules governing the transaction.

Of course, if banks offer to some large customers that submit many payment orders the option of preparing the magnetic tapes themselves, those customers will be aware of the changes in procedures, and some new rules may be called for, since, at a minimum, account must be taken of the impact of errors in the tapes prepared by the customers. However, this problem is easily solved by agreement between the parties. It is not surprising that countries in which such situations arise have some doubts as to whether any preparation of new legal rules is necessary. Nevertheless, even in these countries and in regard to purely domestic transfers, it is becoming clearer that the changes in procedures raise doubts as to whether all of the old rules are still valid. How is a funds transfer to be authenticated? Are the old time limits still valid? When is a funds transfer so complete that a payment order can no longer be revoked, the beneficiary’s bank owes the beneficiary, the underlying obligation is discharged, and the account is or is not subject to legal process?

Check-oriented countries are in a completely different situation. Electronic funds transfers in these countries do not entail the truncation of paper and the sending forward of data by electronic means, as it does in giro countries. Truncation of checks can be done, of course, but the banking and legal problems are quite different. In any case, the use of electronics tends to call for transfers in the nature of giro transfers, known more generally as credit transfers, rather than transfers similar to check collections, otherwise known as debit transfers.

In check-oriented countries, such as the United Kingdom and the United States, there was no complete set of legal rules already in existence to which the new electronic funds transfers could be assimilated. Everything was new. Rules could be, and have been, adopted between the participants and particularly by the clearinghouses. These rules obviously affect non-bank customers, as well as the banks that are not members of the clearinghouses, even though the classical rules of contract law would restrict the effect of the rules to those parties in privity of contract. It may be that in the United States these rules have a general effect binding on all parties to the transaction and not only on those parties dealing with the clearinghouse by reason of special rules in regard to the banking system. But obviously it is desirable to have a complete legal structure that can cover the fundamental banking issues arising in any credit transfer and to do so in light of the use of electronic techniques as well as paper-based techniques. It is not surprising that these two countries have either begun investigations into the appropriate law governing electronic funds transfers or have actually begun drafting a statute.

The United Kingdom and the United States share two other characteristics: each has in place a domestic on-line system for large-value funds transfers in addition to a separate off-line system for bulk low-value funds transfers, and their currencies are major currencies used in foreign trade. As a result, many funds transfers that begin or end in foreign countries, or that both begin and end in foreign countries, pass over these U.K. and U.S. domestic systems and are subjected, at least in part, to U.K. and U.S. domestic law. Naturally, this makes the future law of these two countries of great interest to everyone.

If I have singled out these two countries, it is only because of the characteristics they share, which make their future actions of such great importance to all countries. However, some or all of these factors are present in every banking and legal system. Transfers by the Society for Worldwide Interbank Financial Telecommunication (SWIFT), by proprietary telecommunications systems, or by telex originate and terminate in all countries. The old ways of handling these transfers are no longer sufficient, especially when the funds transfers are international. The banking community has recognized this fact and acted to overcome some of the problems.

At the technical level, the Committee on Banking and Related Financial Services of the International Organization for Standardization (ISO) has been preparing a number of international standards involving, among other things, establishment of agreed terminology and preparation of standard message formats for the different types of electronic funds transfers that can take place internationally. This includes both low-value plastic card-generated funds transfers and higher-value transfers effected by telex or by on-line, computer-to-computer transfers. At least to my knowledge, funds transfers effected by the exchange of magnetic tapes are not made internationally, although there are no technical reasons why they could not be. Of this work of ISO, the most important element for lawyers is the agreement on standard terminology. If we are to have uniform rules for any portion of these transactions, we need to agree on the terminology; and it is preferable for us to use the same terms as are used by the operational personnel in the banks, at least so long as that terminology meets our needs.

A second effort is being undertaken by the Banking Commission of the International Chamber of Commerce (ICC), which has been developing rules to govern certain aspects of funds transfers between two banks in different countries. These rules have a certain legal content to them, but they are very limited in scope. They consider primarily the indemnity due from one bank to another if there has been a problem in the funds transfer, and they restrict their coverage to two banks that are in different countries. These ICC draft rules do not cover a bank-to-bank transaction within one country, even if that transaction is part of a funds transfer that originates in one country and terminates in another; they do not cover the relations between customer and bank at either end of the funds transfer; nor do they cover customer-to-customer relations.

These two current efforts by ISO and the ICC demonstrate the wide interest of the banking community in reducing the problems arising in international funds transfers. The problems the two organizations consider are different, but they have a common antecedent. The introduction of computers into the making of funds transfers has permitted an astronomical increase in the numbers of such funds transfers between different countries and has caused the banking system to abandon the traditional use of checks for that purpose. Standardization efforts, whether in the technical area or in legal relationships, that could previously be limited to the national sphere now must be international in scope.

UNCITRAL now has started a third effort in this field that relies in part on the work already done by ISO and the ICC. It is undertaking the preparation of Model Rules for Electronic Funds Transfers. This effort has a history that goes back some 15 years owing to a desire on the part of the Commission that the increased use of telecommunications in international funds transfers be properly reflected in the development by the Commission of the two draft conventions then before it, one on international bills of exchange and international promissory notes and the other on international checks. I might mention that one direct outcome of that concern was the termination in 1984 of the work on international checks—a project that obviously no longer served a function. We have concentrated instead on the convention on international bills of exchange and international promissory notes—a project of potentially great importance.

By 1979, the UNCITRAL Secretariat and its advisory group in this field, the UNCITRAL Study Group on International Payments, had become sufficiently interested in electronic funds transfers to send a questionnaire to all central banks inquiring into the efforts then being undertaken to introduce electronic-funds-transfer techniques into their payments systems and—what was more important in the long run—as to what the legal rules were that would govern such funds transfers. We received a large number of replies, mostly from central banks but also from some private commercial banks. It is also interesting to note that in a number of cases, the replies were written by the data-processing sections of the banks and not by the lawyers. It seems surprising today, but it is a fact that by and large, the data-processing people did not see that they had changed anything by introducing computers into the funds-transfer process. In their view, all that had happened was that there were new technical tools with which to debit and credit accounts.

The questionnaire was a failure and, being a failure, made it clear that the respondents collectively had no idea whether the paper-based rules continued to apply and whether they were appropriate. This, of course, conformed to the Study Group’s own view of the matters; everything was unclear.

Therefore, the Study Group recommended, and in 1982 the Commission decided, that a legal guide on electronic funds transfers should be prepared that would be oriented toward exploring the impact on the law of the fact that some or all of the steps involved in making funds transfers that had previously been paper-based were now electronic. The Legal Guide was finally published in 1986.

The Legal Guide was an exploratory text. It was not designed to give answers as to what the rules should be, only as to what factors might affect the drafting of appropriate rules. It was thought that this was a prerequisite to the establishment of an international consensus on the content of those legal rules. However, it is obvious that even more than that is really necessary if there is to be consensus on such rules. Therefore, in 1986, in connection with its decision that the Legal Guide should be published as the product of the UNCITRAL Secretariat, the Commission decided to undertake the preparation of the Model Rules on Electronic Funds Transfers. In November 1987, a first meeting was held, at which a number of questions were discussed; and at the end of the meeting, the Secretariat was requested to prepare a first draft.

The first question that might be asked is exactly what we mean by model rules. Perhaps it is ingenuous to say that we do not know ourselves. We do not know to what extent it will be possible to find agreement on some of the important questions that arise in the process of making such funds transfers. Will it be necessary, for example, to draft two or three different rules on a given point and report that these formulations adequately reflect the different positions taken in the world? If this is the result, the Model Rules will serve to clarify the issues, but will not themselves become law.

We may be able to go beyond this and formulate provisions that represent true agreement on a number of important points. We will then be faced with the question whether these rules should remain a form of restatement, should be cast as a model law for adoption by countries, or should be cast as an international convention leading to the unification of law on the points in question. Only time will tell the degree of success we will achieve.

Whatever may be the eventual legal form of the rules, the basic scope of the project remains the same. The Model Rules are not restricted to the bank-to-bank relationship, as are the ISO and ICC efforts, but instead cover the entire customer-to-customer funds transfer, including the bank-to-bank relationship. This appears most clearly in Article 16 of the draft Model Rules, which deals with certain civil consequences of funds transfers.

You may be interested in a number of the basic issues that arose in the preparation of this draft and the solutions proposed by the UNCITRAL Secretariat or, in a few cases, adopted by the Commission or the Working Group. The basic decisions go to scope of application, and these decisions then influence a large number of substantive issues.

As already stated, the Commission decided that the Model Rules should cover the entire funds transfer from originator to beneficiary. This would not be surprising in a domestic effort to draft a law on the subject, but it is less obvious in an international effort. When the originator is a merchant, the payment order it gives to its bank is essentially the same whether the beneficiary is in the same country or in a different country. The nature of the bank’s obligation to its customer would normally be considered to be a matter for domestic law. The same would seem to be true of the relationship between the beneficiary and its bank. Let us take, by way of example, the question as to when the originator’s bank may debit the originator’s account and when the beneficiary’s bank must credit the beneficiary’s account and make the funds available.

If we look at the funds transfer from the viewpoint of the originator and the beneficiary, however, we see that these questions vitally affect the ability of the originator as debtor to discharge a contract obligation to pay the beneficiary as creditor. It is not sufficient to say, as many countries do, that the time the beneficiary’s bank credits the beneficiary’s account is a matter of negotiation between those two parties. If such a rule is to be acceptable, it must be accompanied by a rule that the originator as debtor fulfills its contractual obligation by seeing that an appropriate payment order and adequate cover arrive at the beneficiary’s bank and that what occurs subsequently is of no interest to the originator.

This conclusion, which is the proposal of the UNCITRAL Secretariat in Article 16 of the draft Model Rules, brings us even further than the relationship between banks and their customers at either end of the funds transfer. It also calls for the inclusion, in this text on funds transfer, of a rule on the time monetary obligations are discharged by means of a funds transfer. In my mind, at any rate, the two problems are so closely linked that they must be considered together, even though in many legal systems the problems are thought to fall in entirely different areas of the law.

The use of the term “funds transfer” rather than “electronic funds transfer” brings up a second important issue, what is meant by the term electronic funds transfers in the context of legal rules. My answer to this question is that it depends on the kind of funds transfer for which one is trying to prepare legal rules. If the rules relate to a particular message system, whether or not the system involves a settlement arrangement—and therefore if the rules relate to SWIFT, Clearing House Interbank Payment System (CHIPS), Clearinghouse Automated Payments System (CHAPS), Bankers’ Automated Clearing Services (BACS), or a particular point of sale (POS) system (EFT-POS)—it is not necessary to define what is meant by electronic funds transfers. The rules will be written in terms of the technology used.

Once we broaden the rules to include several existing or potential systems, the problems become more difficult. Therefore, in the United States, for the purposes of the 1978 law,1 there is a definition of electronic funds transfers that is based on the means by which the customer initiates the transaction. That definition works in a consumer-protection law such as the 1978 law. Once the rules are intended to cover the basic law of funds transfers rather than of funds-transfer transactions—that is, once the rules are to govern the series of messages flowing between different banks to implement a single payment order of a customer—it is no longer possible to restrict the rules to electronic funds transfers, however they may be defined. Even between banks, and certainly between customers and banks, some of the payment orders will always be on paper.

A similar conclusion has been reached in the ICC in its drafting effort on interbank payment rules, in the United States in the effort to draft uniform legislation for high-value funds transfers, and in UNCITRAL; the scope of application of the rules must include both paper and electronics—and voice and optics. The substance of the rules must therefore be appropriate for all these technologies, as well as any others that are likely to come into use in the reasonably near future. This is really not as difficult as it may seem at first, as should be evident from a review of the draft Model Rules prepared by the UNCITRAL Secretariat. You may not agree with a number of the specific rules that are proposed, but you will probably agree that few, if any, of them are technology-specific.

If the draft Model Rules are in fact not technology-specific, are they designed to cover all funds transfers? And should they be? The answer in regard to the Model Rules is that at present they are clearly not designed to cover all funds transfers. However, their eventual scope in this respect is still very much open to discussion.

As most of you probably know, an effort was made in the United States beginning about ten years ago to draft a New Payments Code that was designed to cover all forms of payments, even those made in cash. The theory was that since the ultimate effect of any form of payment is the same (i.e., the transfer of monetary value from one part to another), it should be possible to identify the essential elements of a payment and create a set of legal rules that was common to all forms of payment in respect of those essentials and to have special rules for each form of payment only in respect of essential differences between them. The effort failed. Perhaps it was too ambitious; perhaps the politics of the drafting effort were not handled as well as they might have been. Whatever the reasons, the conclusion drawn in the United States was to undertake the preparation of a new legislative text that is restricted to high-value funds transfers and essentially those that go over CHIPS, FedWire, SWIFT, or by similar telecommunications networks.

This decision can also be questioned because it will tend toward the fractionation of the law into a series of legal texts, each one of which governs only a small portion of the total funds-transfer process.

So far, the only firm decision made in UNCITRAL in regard to the Model Rules is that they will not cover negotiable instruments, including checks, or their collection. Any changes in the law called for as a result of the truncation of checks at either the bank or branch of deposit, or at some central place with electronic forwarding for presentment of the payment data, should be made in connection with the specific law governing such instruments.

A broader, but temporary decision is that our Model Rules will not, at present, be drafted to cover debit transfers of any kind. Two reasons have been given for this decision. The first is that credit transfers are operationally easier to describe and, therefore, to write legal rules for. Let us see what level of agreement we can reach internationally on the law governing credit transfers before we try to write rules for electronic debit transfers. The second is that international funds transfers are essentially credit transfers at present. However, international debit transfers do exist, and they will become more common in the future. It can be expected that claims for reimbursement under documentary letters of credit will eventually all be made electronically.

I am not aware of any country that has statutory law on the subject of debit transfers except in regard to negotiable instruments and their collection. Electronic direct-debiting schemes that exist in some countries seem to be based exclusively on contractual arrangements. Since this is so, it may well be that a strong need will be felt for an agreed legal framework for electronic debit transfers. In any case, the current decision that the Model Rules will concentrate on credit transfers will be reviewed at some time in the future.

As one turns to the substantive aspects of the Model Rules, one interesting question that arises is the nature of a receiving bank’s obligation to react to the payment order it has received. It is undoubtedly a universal rule that a bank that receives a payment order from one of its non-bank customers has no obligation to execute the order if the customer does not have sufficient credit to its account and has not made other satisfactory arrangements to reimburse the bank.

The same rule may not apply when the payment order comes from another bank. The rule for some or all domestic transfers may be that the receiving bank is automatically obligated to execute the payment order. Such a rule makes sense in a country where all the banks have been nationalized. There is no question as to whether one nationalized bank will reimburse another nationalized bank. This rule is also in effect in the United Kingdom for CHAPS. The advantage of such a rule is that it speeds the entire payment system. A receiving bank over CHAPS need not—indeed, cannot—wait until the end of the day to see whether the sending bank will settle for the payment order before the receiving bank credits its credit party. Receiving banks bear a credit risk, which runs into millions of pounds daily, that they will not be reimbursed. This risk can be run only because access to CHAPS is limited to large banks that are well funded in pounds sterling. I make the point that the CHAPS participants must be well funded in sterling, since there is one foreign bank among the CHAPS participants. Furthermore, the Bank of England closely supervises the CHAPS participants, thereby giving assurance to all the other CHAPS participants, and there is a widespread belief that in the case of problems the Bank of England would stand behind CHAPS settlements.

It is obvious that no such rule could apply to payment orders received by a bank in one country from a bank in another country. At a minimum, any rule would have to allow the receiving bank to determine whether adequate cover had been offered to it before it was obligated to execute the payment order it had received. That is, the rule in regard to receipt of a payment order from a foreign bank would be the same as or similar to the rule in regard to receipt of a payment order from a non-bank customer.

The Working Group, at its meeting last November, told us that we should not leave the matter at that. They agreed that the Model Rules themselves should not under any circumstances require a receiving bank to execute a payment order. Such requirements could be left to rules of any funds transfer system involved or to the contract between the parties. However, the Model Rules were to require a receiving bank that was not going to execute the order to notify its sender of that fact. This is to protect both the sender and the originator, both of whom expect the funds transfer to be carried out in the normal way. The Working Group also made it clear that this duty to react to a payment order by executing it or by giving notice that it would not do so arose whether or not the sender and receiving bank had a pre-existing correspondent or other contractual relationship. This was thought to be a duty arising out of being a participant in the funds-transfer process.

Interestingly enough, the Working Group specified that the receiving bank that was not going to execute an order should not be required to give a reason for its failure to do so. The Working Group recognized that it could be very important for a sender to know the reason. After all, if the sender is to solve the problem, it needs to know what the problem is in order to do so. However, the Working Group was conscious that the reason for refusing to execute the order could also be that the receiving bank suspected that the funds transfer was a money-laundering operation. It did not feel that the receiving bank should be required to tell the sender of its suspicions. The receiving bank might face one set of problems if its suspicions were correct and an entirely different set of problems if its suspicions were incorrect.

Once there is a rule that a receiving bank has to react to the receipt of a payment order, there has to be a rule as to the period of time within which that reaction has to take place. In some contexts, the rule might be one of instant reaction, such as in CHAPS, as already discussed. In CHIPS, the rule would be that either the execution or the notice of non-execution would have to take place within some short period, measured in hours, after the settlement on the evening of the day the payment order was received. In some countries, it might be thought that the appropriate period within which a bank would have to react to a payment order received by telex should be measured in days, weeks, or even months. The draft prepared by the UNCITRAL Secretariat says the bank must react the same day, but this statement is obviously only a starting point for discussion.

Another question that arises is what a receiving bank should do when it expects to execute a payment order but has not yet received notice of cover. This is not such an easy question. It could await receipt of the notice. Our draft says that when the time for waiting passes the limit—currently the end of the day the payment order was received—the receiving bank must give notice that it is waiting. That is a straightforward application of our general rule. One way out of possible difficulties would be to permit the receiving bank to execute the payment order conditionally, so that it could retract its execution if it never received cover.

The problem is, of course, that we have two competing desires. On the one hand, we want banks to execute payment orders as promptly as possible. The CHAPS rules are the ultimate in this regard. On the other hand, we want certainty in our transactions. We will not have certainty if payment orders can be reversed simply because the sender finally realized it would not receive settlement from its sender. It is easy to construct horror situations with chains of supposedly final funds-transfer transactions being undone because of failures of settlement several banks up in the chain. Not only is this prospect not good for banks, it is not good for merchant customers. Neither the originator nor the beneficiary would know whether the funds transfer was complete and the underlying obligation discharged.

Another major question of interest to banks and customers alike has been the matter of liability when things go wrong. It is fair to say that the typical development of the law recognizes that when an activity is first undertaken to furnish goods or services, that activity is often unreliable. Therefore, the furnishers of the goods or service are able to say to their customers that the customer has to undertake the risk that the goods or service will not work. Few customers place much reliance on their performance. Over time, the good or service is improved, and failures arise less often and are more easily attributable to poor planning or execution. Reliance becomes greater. The old rule exonerating the furnisher of the goods or service from the consequences of failure becomes less justifiable or acceptable. Nevertheless, the beneficiaries of the rule hang onto it, acting as though they would promptly go out of business if it were changed. In their eyes, it has become an acquired right.

This same path has been followed in respect of funds-transfer services. In particular, in earlier days, when international funds transfers relied upon the mail or pouch services to move paper from one country to another, it was obvious that banks could not be held responsible for much that happened outside their own premises. With the advent of electronic transfers, the old uncertainties have been reduced. It is now possible to know how long a funds transfer should take, and the reasons for error, while they have changed in type, have also been reduced in number.

The first draft of the Model Rules, therefore, contains several provisions that would broaden the responsibility and potential liability of banks in comparison with what they have been in the past. The most important of these provisions is that the originator can hold its bank responsible for the proper completion of the funds transfer in an appropriate period of time. As mentioned before, the funds transfer would be completed once the beneficiary’s bank had a proper payment order and cover. What the beneficiary’s bank would do subsequently would be of interest only to it and to its customer, the beneficiary.

The basic obligation of the originator’s bank is, therefore, one of performance of the funds transfer. If it were held liable to the originator for some loss caused from a failure in that performance, it could seek reimbursement from its receiving bank, and so on down the line until a bank was encountered that could not seek reimbursement because its receiving bank had carried out the order it received. In other words, the loss would be left with the bank where the problem had occurred. This obligation is similar to that found in the international carriage of goods, where the shipper of goods can normally hold the carrier with which it has contracted responsible for damage to the goods wherever that damage may have occurred during the voyage. It has the advantage of permitting the customer who has suffered a loss to pursue its claims in its home country. It throws onto the banking system the obligation to find out what went wrong and where. After all, the banking system, taken together, has all the necessary information.

Saying that a bank has undertaken to see that the funds transfer is properly carried out does not mean there are no excuses for non-performance. Without becoming too technical, the draft of the Model Rules, as prepared by the UNCITRAL Secretariat, contains a force majeure clause. You will find in Article 15 two variants. One is essentially taken from the United Nations Convention on Contracts for the International Sale of Goods, a convention that was prepared by UNCITRAL. The other we borrowed from the U.S. draft, and it is more oriented toward funds transfers.

What I find most interesting in this latter draft provision is that computer or communications problems are an excuse for failure to execute a funds transfer only if the problems are not the result of lack of ordinary care by the bank and if the bank has exercised the diligence the circumstances required. In my view, this means that banks must have backup facilities available and disaster-recovery programs in place in order to escape civil liability from their customers. Of course, they ought to have those in place anyway, so this text would not add to their operational burdens. But this text would add to the financial burden of those banks that did not plan for the possibility of computer or communications problems and then tried to leave liability for some of the financial consequences of these with their customers.

As to the level of liability, the primary dispute is over indirect damages. From the viewpoint of bank customers, it is unacceptable to say that they must use the banks to conduct their business but that the banks are not responsible for the losses caused by their failure to act. From the viewpoint of the banks, the threat of having to pay indirect damages is frightening, because it is impossible to estimate what the total of those damages might be and, therefore, what the liability-insurance premium should be. In Article 12, we have offered a suggestion that the bank should be liable for indirect losses, but that there should be a limit on liability equal to the value of the funds transferred. The limit is arbitrary, and in some cases quite high, but it is objective and would provide some basis for calculating insurance premiums. Perhaps this will turn out to be an acceptable idea. If not, perhaps it will stimulate a better idea. In any case, it is an issue that will have to be considered.

There is no doubt in my mind that this draft will go through a number of revisions before the Model Rules are finally adopted. It is not so important whether the current draft already articulates the best rule possible. What is important is that it presents a structure that will permit a rational discussion and that can serve as the basis for the development of the final text. If we have achieved that modest goal, I feel confident that the Model Rules will serve at least to lead to a common understanding of the issues involved and to appropriate ways of addressing those issues. But the time may also be ripe for a much more ambitious achievement—the unification of the basic rules governing international funds transfers.


The Draft UNCITRAL Model Law on International Credit Transfers, which appears in Appendix IV of this volume, is the draft current at the date of publication of this volume. As of that date, this draft was under consideration by the Commission.