Electronic Trading - legal aspects

Professor Olav Torvund, Norwegian Research Center for Computers and Law, (NRCCL) Faculty of law (WWW-page in norwegian only), University of Oslo.

The article has been printed in NRCCL's 25 Years Anniversary Anthology (Jon Bing and Olav Torvund (ed.)), TANO Oslo 1995.




1 Introduction

A major part of international business transactions is handling of information. Traditional international trade essentially consists of moving goods and/or services in one direction, and money in the other. This cannot be done without sharing information between all the parties involved in the transaction, parties such as the manufacturer, agents and brokers, suppliers of transport services, banks, insurers, custom authorities in at least two countries, etc. 50 parties or more may be involved in one trade transaction.

Paper documents represents an old and ineffective technology. If information is communicated on paper, one need to transport a lot of paper in addition to the actual information. It takes time, and it is expensive. The paper is bulky, and it is an insecure technology for storing and transmission of information. For those reasons, there is a strong desire to get rid of paper in trade, transport, finance and administration. One wants to utilize modern informati- on technology for a more effective and rational handling of infor- mation.

There has been written a lot about legal obstacles that can be impediments in the process of introducing new technology. And there has been expressed concerns about the possibility of using computer records as evidence and the probative value of such records. But there are some other legal challenges that we will meet when we get beyond those problems. In this article I will discuss a few of them.

A key issue is how to secure ones rights when the values are transferred or other rights are established, when there is no docu- ments representing those rights. When one will secure the rights to a monetary obligation, title of ownership or other rights, the rules applied can be based on one or more of four principles:

 Possession of the object. That is the basic way of securing one's rights, and it is convenient for movable goods.

 Possession of a symbol representing the object or the right, typically a document. If there are rules saying that possession over the document gives title to the object or the right, the document can be treated as moveable goods.

 Notification to the debtor or the possessor of the object or the one who holds the document representing the rights.

 Registration in a registry. For a long time there has been registries for real estate, for ships, aircrafts etc. in many countries. Over the last ten years, there has also been established registries for securities in several countries, among them Norway.

I use the term registry when the register is held by a third party, and notification when the debtor or someone acting on his behalf is notified. But there is no sharp division. The registry will of course be notified, and a debtor with more than a very few obligation will keep a register. And when someone is notified on behalf of the debtor, one can put it into both categories.

Traditionally, documents was easy to hand over, but it was a complicated matter to register one's right in some registry with bound books and paper dossiers. For those reasons, documents has been preferred for rights in objects that were moved around, were traded in frequently etc, while registries was preferred for rights in goods of high value that were not moved too much around and not traded in too frequently.

Modern information technology has turned the situation upside-down. Documents has to be delivered physically, and that is not very convenient when the buyer and seller is far away from each other. Transportation of documents is slow, expensive and insecure. Telecommunication is faster, cheaper and gives better security. One can add that high volume of transactions in many markets makes it practically impossible to maintain a paper-based system.

Today we are moving from systems based on possession of documents to systems based on notification or registration. Modern telecommunication makes notification easy, and one can easily maintain large book-entry systems.

There are also some hybrid systems: The documents are stored by one depositary. The depositary keeps a register, showing ownership and other rights in the documents, and the depositary is notified about transfers etc. The main advantage of the a hybrid system, is that one can maintain the old documentary rules, making the transition more easy. But the systems will have well established rules for transactions that do not take place anymore(transfer of documents), and no legislation for the book-entry system estab- lished on top of the paper-based system. And one will still have the problem of secure storage and from time to time transport of paper.

CMI has established rules for ®Electronic Bill of Lading¯. The title is misleading: It is not rules for an electronic bill of lading, it is an electronic alternative to using bills of lading. The system is based on notification to the ship carrying the cargo. In funds transfer and in part of the securities market, there are also systems based on notification.

A system based on notification might be convenient when only the rights of third parties are concerned. But if the information held also relates to the content of the obligation, it has clear weaknesses. A bill of lading is both a contract between the shipper and the carrier, and a document of title to the cargo. The bill of lading not only tells who has the right to receive the cargo, it also tells what the carrier shall deliver.


2 Some legal consequences

When analyzing the legal consequences, one should distinguish between the consequences of leaving the old document-based systems, and the consequences of establishing the new systems.

The document-rules might be divided into three categories, al- though there are no clear distinctions between them.

The first category consists of rules making the use of certain documents mandatory for specific transactions. In Norway, every limited company had to issue share certificates until the legislation was changed in 1985. It was not possible to establish a paper-less system for securities as long as we had those rules, and the rules was amended to facilitate for a paper-less securities registry.

This kind of rules can stop the introduction of EDI and paper- less systems. But if there is a willingness to introduce electronic systems, rules can be changed. If one analyzes the legal environment as part of the planning for an electronic systems, rules of this kind are easy to identify. If there are rules like that, the process might take some more time, and one will need political support. But from a legal point of view, it is not a very difficult situation. (It might of course be difficult if one is trying to find a way around such rules, but I am not going to discuss that aspect.)

The second category is rules making the use of a document a condition for validity or enforceability. The rules might require the use of a certain document with a specific content, signatures, witnesses etc, or it might just be a requirement for a document in writing. In many countries one can find such requirements for contracts regarding real estate. In Norway, writing is not a require- ment for such a contract to be valid or enforceable. But to protect your title or any other right in real property, the right needs to be registered in a traditional registry of land. And one have to present a written and confirmed document for the registrar.

The third category is where a set of rules are triggered by the use of a certain document. There are no rules saying that one have to use a bill of lading, a bill of exchange, a cheque, etc. But if you use any of those documents, there are certain rules that applies to the transaction.

A comparison of a manual debitcard transaction and a cheque transaction will illustrate the point. If one looks at the flow of information and what the involved parties are actually doing, the transactions are almost identical. If the card-issuers had printed the magic word ®cheque¯ on their paper-slips, a debitcard transaction would have been a cheque transaction. But from a legal point of view, they are very different: The cheque transaction is governed by legislation, and for a number of countries (not UK and US) by the Geneva convention of 1931. The debitcard transaction is in most countries governed only by contracts and general principles of law.

A variation of the requirement for a certain document, is a requirement for a transaction to be in writing, and/or a requirement for a signature. The legal issue here will be if those requirements can be met in an electronic system. It will depend on the interpretation of the statute. If writing is required for the purpose of evidence and notoriety, the requirement will probably be met by an EDI-message. If on the other hand the reason is the need for an original, as is the case with negotiable instruments, it is unlikely that an EDI-message as such will meet the requirement.

There has been some attempts to give general definitions of ®writing¯ and ®signature¯, which will include EDI-messages and electronic signatures. As the requirement for writing and/or signature has different purposes in different transactions, I do not think that such general definitions can be successful.

The Vienna sales convention art. 13 says that ®writing¯ includes telegram and telex. The convention is from 1980, and EDI was not an issue when the convention was drafted. Today one would certainly have said that ®writing¯ also should include EDI- messages. Even with today's wording, an EDI-message will probably be regarded as a written message in relation to the convention. A telex is after all a telecommunicated message printed out locally by the receiver.

When it is a question regarding a negotiable instrument, it is less likely that the wording ®in writing¯ will be given a broad interpretation. The cheque legislation requires the document to be in writing, and the document has to be presented for the drawee. But these requirements were at least partly set aside by the Norwegian Supreme Court in a decision published in Rt. 1991 p. 1335. The cheque was deemed as presented when the relevant information was presented in electronic form, the actual document was never presented to the drawee.

Rules in the third category will not represent legal obstacles in a process of introducing paper-less electronic systems. But if we move toward a paper-less system, we will leave the well known and tried rules and enter into unexplored legal territory. One will have legal uncertainty and a higher risk.

I have frequently been asked if there are legal obstacles to introducing EDI in certain areas of business. The answer to that question is generally no, at least under norwegian law. But it is not the right question to ask. The important and interesting question to ask, and the difficult one to answer, is what the consequences will be if one leave the traditional paper-based systems, and introduce new and yet unregulated systems.


3 From two-party to three-party transactions

Most paper-based transactions are basically two-party transactions. Documents can be handed over, and the transaction can be settled between two parties. The most obvious example is probably payment. A monetary obligation can be liquidated by a direct delivery of cash. When using any other payment instruments, a bank will be involved in the transaction.

Third-party registries are being introduced in many new areas, and the involved parties will often not be aware of that change in their way of doing business. The system is established for technical and practical reasons. One need someone to manage the flow of information. The content and the function of that information is quite often not analyzed, and one has not analyzed the legal conse- quences of leaving that information to a third party. The third party involved is frequently mentioned as a ®network provider¯, or as a provider of ®value added network services¯.

One have to distinguish between three-party systems where the third party is a party to the transaction, and systems where the third party is only keeping information about the transaction without being directly involved. A bank is a party to a funds- transfer transaction. The bank does not register the debit or credit balance directly between the payer and the beneficiary, but the banks position towards the two parties. The transaction is actually split into two two-party transactions, with the bank being a party in both. In many markets for derivative financial instruments, one can find the same concept.

A system where the third party is not a party to the transaction, will in it's simples form just record traffic. One can register messages sent and received, without being involved with the content or the processing of the message. It will be an electronic equivalent of registered mail. But even on that simple level, there will often be authentication mechanisms and other securities measures, formal checking, etc provided by the third party. There might also be a conversion from one message-format to another.

In this article I will not discuss the third party's liability when providing such services. But the one who takes responsibility for the security in the system, will run the risk of being held liable for a substantial amount of money if the security fails.

The next step is to store the messages for a certain amount of time. The third party might do that in his own interest, as a way to keep evidence of the service that has been provided. A storage of that kind rises several data-protection issues that I am not going to discuss. But that information will also be of interest to the other two parties in a transaction: It will be a valuable source of evi- dence if there should be a conflict.

The banking network S.W.I.F.T. stores all messages in encrypted form for a period of 4 months. They do it for their own purpose. But this information is frequently asked for by the parties using the S.W.I.F.T. network. They need that information to settle disputes where S.W.I.F.T. is not a party.

The concept can be developed further. All the parties using the system can accept to be bound by the information in the registry. In principle the can be agreed between the parties using the third- party service, without involving the third party in their agreement. But the practical way will be to make this part of the contract one enters into when subscribing to the service. There has been made several attempts to establish such systems, and SeaDocks is proba- bly the most famous example. The problem so far has been uncer- tainty regarding liability, and lack of financial support.

One major weakness of a system based on contracts, is that it will not be binding for third parties who has not entered into the contract. If my understanding of UK law is correct, the privity rule makes it difficult even to bind parties who has a contract with the system.

The final level is registries based on statutory regulation.

Both third-party registries and notification are legal techniques with long traditions. One might therefor say that they do not represent anything new. But they are introduced in new areas, and they are not restricted by limitations and lack of flexibility in old technology. In old registries, one had to choose one and only one key for sorting information: One could either sort by subject or by object, but one could not do both. With modern databases, one can use many different indexes, and even free text searches. And telecommunication makes it unimportant where the registry is located.

What is really going on, is that the commercial infrastructure is being reorganized. New services are needed, and new parties are providing those services. These new services has no traditions, and no established legal framework. New services will to a large extent be provided by new organizations. On the other hand will other services with long traditions not be needed anymore. If existing organizations is going to survive, will depend on their ability to adapt to the new situation.

The developments in the securities market illustrates the situation. A stock exchange, commodity exchange or any other exchange used to be a place where buyers and sellers met in person. In a modern world, a computer system can be a much more efficient marketplace that can take hand of both dealing and settlement. 14 days after ®The Big Bang¯ in London, the trading floor at The London Stock Exchange was virtually empty. It did not make sense to keep a large trading floor in City of London where office space is not known to be cheap, so the trading floor was closed. In the old days, the trading floor was the stock exchange. Nobody needs the old exchanges anymore. But they have adapted to a new situation. Now the exchanges are running the computer systems, and they supervise the market. Such development are not without conflicts. There are other surveillance bodies, other organizations that want to provide computer and network services etc.

The legal challenges does not come from the technology as such, but from the changes in business and administration made possible by new technology. The impact of those changes will vary from sector to sector, and one cannot give a general answer to the challenges. But there are some tendencies that will give some of us lawyers many challenges in the years to come. I will discuss centralizing, decentralizing and internationalization.


4 Centralizing, decentralizing and internationalization

Use of telecommunication means that two parties dealing with each other do not have to be at the same place. If the transaction is registered in a registry, they do not have to be at the same place as the registry. And if someone is seeking information from the registry, they can access that information via telecommunication.

When using paper, the parties had to meet somewhere to exchange paper. Brokers had to be located close to each other and close to the exchange, and clearing banks had to be located close to the central bank. They don't have to anymore. It is now possible to sit at a beach somewhere, link up a laptop computer via mobile telephone, and deal with people sitting on another beach somewhere else in the world. If the transaction have to be registered somewhere, or someone have to be notified, the necessary information may be sent via the same computer. By using mobile satellite communication, one could sit in the middle of a dessert or in a boat in the middle of the Atlantic ocean, and have access to the same information and to the same dealing network as someone sitting in City of London.

Parties who are dealing with each other can be anywhere, so that part of the business can really be decentralized.

Traditional registries where local registries. One had to submit documents to the registrar, and anyone seeking information had to go to the registry's office and look through the files. We do not need local registries anymore. At least in Norway, there is a clear move from local to national registries. For some purposes local registries are maintained, mainly due to tradition. But even where local registry's offices are maintained, the information is transmitted to a central database, meaning that the users can obtain information collected i all registry's offices from one central database.

When one is looking at the registries, the tendency is centralizing. When registries are introduced for transactions where one used to exchange paper documents, there are no tradition of local registries. For these transactions, one will build central registries from the very beginning.

As far as domestic transactions are concerned, the tendency of both decentralizing and centralizing at the same time, does not give very specific legal problems. But decentralizing of the dealing means that it is more likely to have the parties in different countries. A broker in Oslo and a broker in Milan may sit in their own offices and deal with each other in the London market. And if they are dealing in danish bonds, the transaction has to be settled and registered in Denmark.

If we compare this situation with ®the old world¯, there are some very interesting differences. When brokers in the London market had to meet at the London Stock Exchange, it was obvious that these transactions was governed by UK law. And as long as they were dealing in paper, the transaction could be settled by delivery of paper documents in London in exchange with money. As long as the delivery would take place in London, the whole transaction would be governed by UK law.

When the parties to the transaction happens to be in Italy and Norway, it is no longer obvious that it will be a UK transaction. As far as the contractual obligations between the parties are concerned, they might choose UK law by contract. If they are dealing through a UK system, the system rules that the parties has to accept, will usually state that the transaction is governed by UK law. But that will only solve part of the problem.

When it comes to settlement, contract terms cannot overrule danish legislation. The settlement and the ®delivery¯ will in our example be governed by danish law, no matter where the parties are and the law that is chosen to govern the contract.

If the rules governing settlement, protection of the buyers title etc. are harmonized, the internationalization does not cause many problems. But in this new area of law, rules are not harmonized. There are no international tradition or model from which one can derive national rules.

In the scandinavian countries there are harmonized rules governing paperbased bonds. When Denmark in 1983, as the first country, enacted legislation governing paperless bond transactions, they did not want to amend the existing rules governing bond transactions because of the harmonized scandinavian legislation. Instead they drafted new legislation for paperless transactions that was not and could not be in harmony with the old legislation.

When Norway enacted similar legislation in 1985, one could draw on the danish experience, and avoid some of the solutions that had turned out not to work so well in Denmark. Norway too wanted to maintain the harmonized rules for paperbased bonds and enacted new legislation, but a legislation that was slightly different from the danish legislation. And a few years later Sweden has chosen a third solution. The market is becoming more and more international, national rules of the country where the securities are issued are becoming more and more important to the transaction, but at the same time the national rules are less in harmony.

A few examples from funds transfer can illustrate some of the legal issues that might arise from the combination of decentralization, centralization and internationalization. The banking network S.W.I.F.T.has it's headquarter and main operation center outside Brussels in Belgium. It is organized under and governed by belgian law. The Central Bank of France has established a system for large value funds transfers where they use the S.W.I.F.T. network for message transmission. The french system is mainly used for domestic french interbank transactions, although it is also used for international french franc transactions. But every message is sent from France to Brussels, and back to France. And the message handling is governed by belgian law.

For some reasons which I do not know, messages from some northern european countries, among them Norway, are not sent via Brussels. Messages from these countries are sent via the S.W.I.F.T. operation in Chicago. I do not know if the Chicago center is governed by belgian or US law. Once again it seems that the traffic has been moved from Belgium to US for some practical reasons, and nobody seems to have been concerned about the legal consequences. The Central Bank of Norway is working with a large value funds transfer system similar to the french system, where S.W.I.F.T. messages will be used. It would be interesting to know if the handling of the messages will be governed by US or belgian law, but I have not seen any discussion on that particular issue.

One should not exaggerate the importance of message handling, although it illustrate the problem. S.W.I.F.T. is not a party to the underlying financial transaction. The ®money¯ does not move to Brussels or Chicago. But when it comes to clearing and settlement, then the clearing house and the settlement bank will be party to the transactions. For reasons that I will not discuss here, clearing and settlement will always take place in the country where the currency in use are denominated, at least when it comes to large value transactions. The settlement bank will be the central bank in the country of the currency. Transactions in US dollars will be cleared and settled in New York, even if it is a transfer between two norwegian banks. In this case, clearing and settlement will be governed by US law.

Two UK cases, Libyan Arab Foreign Bank v. Bankers Trust Co, 1. Lloyd's 259 [1988] and Libyan Arab Foreign Bank v. Manufacturer Hanover Trust 1 Lloyd's Rep 608 [1989], give us two examples of the rather unpleasant surprises one can get when an essential part of the transaction takes place in a third country and is governed by the law of that third country. The libyan bank had some US dollar deposits in the London branches of two US banks. When president Ronald Reagan issued an executive order, saying that all libyan deposits in US banks should be frozen, the problem was if the deposits in the London branches were governed by the order or not.

The contracts between the banks were rather complicated, involving both New York and London branches of the US banks. The court decided that the contracts was governed partly by UK law, partly by US law. The situation was that a fulfillment of the contractual obligation would be illegal under US law, but not to do so would be a breach of contract under UK law.

From US one accepted that banking activities in London was governed by UK law, even if it was done by branches of US banks. But the US dollar clearing had to take place in New York, and was governed by US law. And one would not accept any clearing of libyan deposits through the New York clearing operation. With deposits of 200 and 300 million USD, payment in cash in London was not an issue. The UK courts decided that the US banks had to pay, but for the issue discussed in this article, the result is not important. The point is that the transactions was partly governed by the law of a third country, and it caused some rather serious and unexpected problems.


5 Conclusions

The major legal challenges are not related to the technology as such. But the utilization of new technology changes the way business and administration being are done. The existing rules do not fit the new business infrastructure. Commercial and financial transactions will be transborder transactions more often than we are used to.

One has to analyze from a legal point of view the new services that are being provided, and the organizations that are providing these services. It is necessary to analyze each type of business or administration, and some crucial legal issues.

A general approach discussing ®Third party services¯, ®Legal aspects of EDI¯, or whatever one choose to call the issue, will only give us some general and superficial overviews. The legal problems related to funds transfer, securities, real estate, shipping, customs, social securities, etc. are not the same, even if they are using the same technology. Mime would ever dream of analyzing ®the legal aspects of paper¯, with ambitions to end up with solutions that are applicable to all kinds of business and administration were paper is being used. Traditional mail has been used in business for ages, but postal law has never been regarded as an important part of commercial law. If the messages are being sent electronically, law related to the message as such will still not be very important to the transaction.


LillOlav Return to my english home-page