Much of the legal status of electronic communications in Canada (and elsewhere) rests on legislation based on the United Nations Model Law on Electronic Commerce of 1996. The Model Law’s main Canadian implementation has been through the Uniform Electronic Commerce Act, adopted in 1999. All the common law provinces, Yukon and Nunavut have enacted the Uniform Act, as shown here. Quebec adopted its Act to establish a legal framework for information technology in 2001, mainly based on the principles of the Model Law though not using the Uniform Act as its template. The electronic documents part of the federal government’s Personal Information Protection and Electronic Documents Act (PIPEDA) was inspired by the draft of the Uniform Act available in mid-1998 when the federal statute was prepared, though some significant variations were made.
The purpose of the Model Law and the Uniform Act, and thus of most Canadian legislation, was to remove barriers to the legally effective use of electronic communications. They did not intend to regulate that use. They wanted to be sure that legal rules that appeared to require ink and paper, such as writing and signature requirements, did not stand in the way of electronic communications. They did so by setting out a ‘functional equivalent’ to the existing form requirements. Electronic communications that performed the specified function were taken to satisfy those form requirements.
These were new principles in the mid-1990s when UNCITRAL developed them. People were not sure whether they would work, and thus how widely it was safe to apply them. As a result of that caution, the Model Law allows for exceptions to the operation of its rules, though it does not say what those exceptions should be. Thus articles 6 (on writing requirements), 7 (on signatures) and 8 (on originals) all have a paragraph reading ‘The provisions of this article do not apply to …’. Each implementing country was left to decide the scope of the rule and what to carve out of it.
In deciding what to include or leave out, Canada had to consider the two main features of the Model Law that were adopted by the Uniform Act: they were minimalist and they were technology-neutral. They set out their rules for functional equivalence but did not say how to achieve them or what technology to apply to do so. (Quebec’s statute was less minimalist but still technology neutral. Parts of PIPEDA are not technology neutral.) Parties to electronic transactions were left to decide what was prudent. Just as a contract written on a Kleenex tissue in pencil, signed with an X, can be a legally effective contract but is unlikely to be acceptable in commercial practice, so too certain kind of electronic communications will not satisfy basic business considerations. The legislation does not make that choice for the users.
With such considerations in mind, the Uniform Law Conference excluded from the Uniform Act a number of communications: wills, testamentary trusts, personal powers of attorney and land transfers that would require registration to be effective against third parties. These documents shared the characteristic that they are often prepared by people without legal advice, people who may have little knowledge of what makes an e-document secure. The decision about prudent practice might be particularly risky to leave open for those people.
In addition, not everyone might appreciate the risks of buying land without registration. Most provinces at the time of the Uniform Act did not have electronic land registration, and Ontario’s system was tightly controlled as to access and technology. Thus whatever people might come up with to transfer land would probably not be registrable in any event. The Uniform Law Conference’s decision to exclude these land transfers reflected the same decision made in then-recent statutes in Australia (see for example the regulations under the Electronic Transactions Act, 2000 of New South Wales) and Singapore (see s. 4 and Schedule 1 of the Electronic Transactions Act.)
As noted in my last column here, the Uniform Act also excluded negotiable instruments, for other reasons.
Recently a number of people have been asking whether the real estate exclusion is still needed, if it ever was. Should the Uniform Act and its provincial enactments be amended to remove that exception, so the legislation would apply to real estate transfers? I will tell you now that I am not going to answer that question; I want to hear your answer. Here are some of the considerations that appear to me to be relevant to the question.
First, we need to appreciate the limited scope of the ‘exception’. The Uniform Act says that where the law requires writing or an original, the requirement may be satisfied by an electronic document of certain characteristics. A signature requirement may always be met by an electronic signature. Thus if the law does not require any of these forms, the legal effect of electronic communications does not rely on the legislation based on the Uniform Act.
Our law rarely requires writing or signatures. As a matter of business prudence, people do tend to ‘get it in writing’, and signatures are widely used for authentication and for the ceremonial purpose of making people realize they are doing something with serious consequences. Just as business prudence leads people to do this, business prudence can govern whether and how they do it electronically.
The main law that requires writing in real estate transactions is the Statute of Frauds. It says that an interest in land needs to be made or created in writing signed by the parties. That statute was certainly present in the minds of the drafters of the Uniform Act. (I chaired the working group of the Uniform Law Conference that prepared it.) The Statute of Frauds applies to the creation of the interest in land, but not to the negotiation that leads to that creation. None of the surrounding documents and correspondence needs to be in writing.
The second consideration deals with the nature of ‘writing’. When the Model Law and the Uniform Act were drafted, it was at least an open question whether electronic communications could be considered to be ‘writing’ for legal purposes. Both documents were created on the assumption that ‘writing’ meant ‘on paper’ (or some other tangible medium), and an electronic communication was not in writing. Given the uncertainty on that point, it made sense to remove the doubt by spelling things out.
Since that time, however, the world has become much more familiar with electronic communications. Courts in several common law countries have held that electronic communications satisfy the Statute of Frauds writing and signature requirements without any statutory help. The Alberta Court of Queen’s Bench in Leoppky v Meston 2008 ABQB 45 found a series of emails capable of constituting an enforceable transfer of a house (though on the evidence, intention to convey was not present). The Court did not mention Alberta’s Electronic Transactions Act, which excludes land transfers in any event (section 7(e)).
The High Court of Singapore decided in SMI Integrated Transware v Schenker Singapore  SGHC 58,  2 S.L.R. 651 (P.C.) that email headers could constitute signatures so as to satisfy Singapore’s Statute of Frauds, despite the exclusion of land transfers (including a lease of the kind in issue) from the Electronic Transactions Act.
Earlier this year, the Commercial Court of the Queen’s Bench Division of the English and Welsh High Court found a guarantee constituted by a chain of emails presumptively enforceable under the English Statute of Frauds. Golden Ocean Group v Salgaocar Mining Industries  EWHC 56 (Comm). The parties agreed that an email signature would satisfy the statute (para. 95).
The third consideration is that not all Canadian jurisdictions have excluded real estate transfers. New Brunswick did not put any exclusions into its Electronic Transactions Act. Quebec’s legal framework statute does not mention immovable property. The operation of the law on this point in those provinces has not created problems, to my knowledge. The American equivalent to the Uniform Act, the Uniform Electronic Transactions Act, did not exclude land transfers. It left the question of registrability to the recorders of land transactions.
For that matter, Manitoba has never proclaimed in force the part of The Electronic Commerce and Information Act authorizing electronic equivalents of writing and signature. So all electronic transactions, not just ones involving real estate, have to find their legal support outside the enabling statute, without noticeable harm to business in that province.
UNCITRAL revisited the question of the legal effect of e-communications in the Electronic Communications Convention in 2005. Confidence in the use of such communications, and in UNCITRAL’s ‘solution’ for removing barriers to them, had grown to the extent that the Convention makes binding on member states what the Model Law had simply put out for adoption or adaptation. The Convention spells out the exceptions, too, as the Model Law had not done. No exception was given, or suggested in the Working Group’s discussions leading up to the Convention, for transactions involving land.
The Uniform Act and its enactments say that it yields to any other law that authorizes, prohibits or regulates electronic communications (section 2(5)). The purpose of this provision was to avoid conflict between the generic permission of the UECA and any existing laws that already created a statutory regime for e-documents or e-filing. The provision clears the way for electronic registration statutes, and ensures that even a general permission to satisfy writing and signature requirements of land transfers would stop short of making an e-transfer registrable.
For example, Ontario’s Land Registration Reform Act, 1994 provides for e-registration of land transfers by special instruction from an authorized user of the system (generally a lawyer for one of the parties) but not for the registration of the transfer documents. Those transfer documents should be legally enforceable in any event, however, if only to justify the lawyer’s instructions to the land titles registry. Thus the e-commerce legislation might still be usefully invoked for the transfer (if it were not excluded), even though the registration is separately dealt with. (One may compare Ontario’s system with that in force in British Columbia, about to become mandatory. As in Ontario, BC’s Electronic Transactions Act excludes land transfers from its general enabling rules.)
(Most of Part 2 of PIPEDA applies only to statutes and regulations specially designated under the Act. It is interesting that the only federal legislation to which it applies, eleven years after it came into force, is the Federal Real Property and Federal Immovables Act. There may be some real demand for electronic land transfers.)
The Uniform Act was clear in its policy that excluding land transfers was not a statement that these transfers should never be done electronically, but only that additional security might be needed. “They seem to require more detailed rules, or more safeguards for their users, than can be established by a general purpose statute like this one.” See the annotation to s. 2 of the Uniform Act.
The practice seems to have evolved since then. Real estate documents are frequently exchanged by fax, and closings done on the basis of faxed documents. Yet faxes are essentially electronic documents, and a faxed signature is no more reliable than a digitized signature that reproduces the handwriting by electronic means. Banks are known to close mortgage financing on the basis of documents in Portable Document Format (PDF), often with signed paper documents to follow – but the money changes hands on the strength of the electronic versions (and no doubt on the presence of considerable other legally enforceable security).
Most land transfers are not done by people all on their own. If they do not have lawyers, they generally rely on at least one and often two real estate agents, who are licensed professionals entirely capable of watching out for basic questions of security, and who of course know about the need for registration of interests being created. The risk of fraud, say by the creation of multiple inconsistent transfer agreements, is arguably no higher for electronic than for paper documents. (The various incidents of mortgage fraud in recent years have not depended on electronic documents in particular.)
In the light of these considerations of law, policy and practice, is the exclusion of real estate transfers from the Uniform Act and its provincial enactments still justified? Given the case law, one might think that the Act could simply be ignored; the Statute of Frauds seems to be comfortable with e-communications – though only one Canadian trial-level court has so held, so far as I know. However, the exclusion is still a message, not a prohibition but a warning, a special status. Is it time for that to change?