The ACCC commenced proceedings on 28 August 2014 in the Federal Court of Australia against Valve Corporation, an e-commerce business based in the USA with no physical presence in Australia, alleging that Valve engaged in misleading and deceptive conduct when it represented to Australian consumers that the statutory consumer guarantees created under the Australian Consumer Law (ACL) did not apply to products supplied by Valve.
It is clearly established in Australian law that the prohibition against corporations engaging in misleading and deceptive conduct applies extraterritorially to statements transmitted over the internet to persons in Australia – see, for example, ACCC v Chen and ACCC v Sensaslim.
But the application of the consumer guarantees (or the non-excludable implied conditions and warranties which pre-dated the commencement of the ACL) to the supply of products to consumers in Australia by foreign corporations that do not carry on business in Australia is a topic of greater legal uncertainty. Whilst there is little doubt that the relevant consumer guarantee provisions of the ACL apply extraterritorially (see sections 6(2)(a), 6(2)(c) and 6(3) of the Competition and Consumer Act), the construction of the provisions of the ACL so far as they apply to cross-border consumer transactions is unclear.
The ACCC’s case of misleading and deceptive conduct against Valve hinges on the application of the consumer guarantees to the supply of allegedly defective software. Each of the allegedly misleading representations pleaded by the ACCC against Valve assumes that the consumer guarantees apply. If they do not, then the ACCC will not have made good its case of misleading and deceptive conduct.
Given the growth of cross-border consumer e-commerce in Australia, the case is of some significance and should be followed by all non-Australian e-tailers with a material customer base in Australia. This blog – which is longer and denser than our typical blogs – examines the relevant provisions and highlights a potential defence that Valve may seek to rely on, if it is not able to settle the proceedings on satisfactory terms.
Some history and context
Prior to the introduction of the ACL, the Trade Practices Act (TPA) operated to imply into contracts for the supply of goods and services to consumers in Australia various conditions and warranties, such as a condition that goods would be of merchantable quality. If there was a breach of such an implied condition or warranty, the consumer would be entitled to sue for breach of contract. The legislation provided that any attempt to modify, exclude or restrict the implication of terms was void (section 68 of the TPA).
The TPA went further and provided (in section 67(a)) that the relevant conditions and warranties would continue to be implied even if the contract specified a governing law other than the law of an Australian jurisdiction so long as the proper law of the contract (ignoring the express term specifying the governing law) would have been the law of an Australian jurisdiction. For this reason, it was necessary to apply Australian common law principles of conflicts of law to determine the proper law of the relevant contract in order to determine whether the conditions and warranties would be implied into the contract. Those common law principles require the court to identify the system of law with which the contract has the closest connection. In the context of a contract for the supply of goods, Australian courts have held[1] that various factors must be taken into account, none of which is conclusive, including:
(a) the place the contract was made;
(b) the place of performance of contractual obligations;
(c) the currency nominated for, and the place of, payments under the contract, and
(d) the place of origin of the goods.
The fact that the products are to be consumed by a person in Australia does not automatically mean that the proper law of a contract to supply those products is the law of an Australian jurisdiction.
The final piece in the puzzle was section 67(b) which provided that if a term of a contract purported to substitute the provisions of the law of a country other than Australia for the non-excludable implied conditions and warranties, the provisions of the Act which operate to imply the conditions and warranties continue to have effect. It is difficult to see why section 67(b) was necessary – any term purporting to substitute the provisions of the law of another jurisdiction for the non-excludable implied conditions and warranties would surely have been inconsistent with those implied conditions and warranties and, therefore, void by operation of section 68.
There is only one case to have considered section 67 in any detail – the NSW Court of Appeal’s decision in Laminex (Aust) Pty Limited v Coe Manufacturing Co [1999] NSWCA 370. The proper law of the contract was found to be the law of the State of Oregon, USA, so section 67(a) did not apply. The decisions of the trial judge and of the Court of Appeal are silent on the operation of section 68 of the TPA, even though the contract contained a term which purported to exclude the implication of any conditions or warranties, including as to merchantability. After providing a limited express warranty, the contract contained the following:
“THE FOREGOING WARRANTY IS EXCLUSIVE AND IN LIEU OF ALL OTHER WARRANTIES WRITTEN OR ORAL EXPRESS OR IMPLIED INCLUDING BUT NOT LIMITED TO ANY WARRANTY OR MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE”
It is not clear whether the court’s attention was simply not drawn to section 68 or if the parties or the court proceeded on the unspoken assumption that section 68 was irrelevant because it did not apply to contracts where the proper law was not Australian law. In the first instance decision, the judge concluded that:
“Section 67 of the Trade Practices Act does not assist [the purported Australian consumer], because I do not consider that … the proper law of the contract under which the goods were supplied would be the law of some part of Australia”
On appeal, the court turned its attention to section 67(b), which did not depend for its operation on the proper law of the contract being Australian law. The ultimate issue for the court was not the effect of section 67(b) or whether the TPA operated to imply non-excludable conditions and warranties into the contract. Rather, it was to decide whether the Supreme Court of NSW should stay proceedings commenced by Laminex in Australia against the US supplier. The Court of Appeal granted a stay, but on condition that if Laminex commenced proceedings in Oregon, the US supplier permit Laminex’s claims based on the operation of the TPA to be tried on their merits. If Laminex pursued a claim in Oregon but the US supplier did not permit the TPA claims to be determined on their merits, Laminex would have been free to pursue their proceedings in the Supreme Court of NSW.
In reaching that conclusion, the court observed that section 67(b) “does create a problem” since the contract at issue contained the term set out above and an additional term purporting to impose a 2 year limitation period for any claims against the supplier. My own view is that those terms were clearly inconsistent with the provisions of the TPA (and, hence, potentially void by operation of section 68) but did not purport to “substitute” the provisions of the law of a different jurisdiction (eg the Oregon Uniform Commercial Code) for the terms implied by operation of the TPA.
The Insurance Contracts Act contains provisions that are similar to sections 67(a) and 68 of the TPA, but there is no equivalent to section 67(b). Those provisions have been the subject of litigation in the High Court of Australia – see Akai v People’s Insurance. The majority in that decision said:
“Taken together, ss52 and 8 [of the Insurance Contracts Act] manifest a legislative intent not only that there should be no power to contract out of the provisions of the Act, but also that the regime established by the Act should be respected as regards contracts the proper law of which is, or but for selection of another law would be, that of a State or Territory. This defeats evasion of the legislative regime by the choice of some other body of law as the governing law of the contract.”
Consumer protection laws of some of the Australian states have contained provisions similar to section 67 of the TPA (but not s68). For example, s17(3) of the Contracts Review Act 1980 (NSW). State courts have held that the NSW Contracts Review Act did not apply to contracts the proper law of which, objectively ascertained, was not the law of NSW: see for example State Bank of NSW v Sullivan and Janesland Holdings Pty Ltd v Simon.
Now we have the ACL
Things changed when the ACL took effect in 2011. In place of the old implied conditions and warranties, statutory consumer guarantees were introduced. These guarantees are, in many ways, similar to the implied conditions and warranties – for example, there is a guarantee that goods will be of an “acceptable quality” which is very much like the old implied condition of merchantable quality. But the remedies for breach of the statutory guarantees are different – no longer does a consumer have a right to sue for breach of contract, but instead has a statutory right to take action under the ACL. Like the old regime, section 64 of the ACL provides that any term of a contract which attempts to modify, exclude or restrict the operation of the consumer guarantees is void – this is the equivalent to section 68 of the TPA. It is qualified by section 64A of the ACL which provides that a term of a contract that modifies, excludes or restricts the consumer guarantees is not void to the extent that it limits the supplier’s liability for contravention of the guarantees in particular ways. The apparent intention of sections 64 and 64A is to allow a limited degree of party autonomy under which the parties to a contract can agree to restrict the commercial impact of a breach of a consumer guarantee.
Section 67 of the ACL is in almost identical terms to the old section 67 of the TPA. It is not entirely clear why section 67 of the ACL exists at all. Now that a consumer’s cause of action is under the ACL rather than in breach of contract it is difficult to understand why the governing law of the contract should be relevant to the application of the statutory guarantee. It was relevant under the old regime because the governing law of a contract determines the rules under which a breach of contract case would succeed and foreign legal systems are unlikely to recognise conditions and warranties implied as a result of an Australian statute. But with the change to the ACL model, either the guarantees (and the related remedies) should apply by operation of the statute or not. If they do apply, then section 64 of the ACL ought to be sufficient to render void any term of a contract which purports to exclude, restrict or modify the operation of the guarantee or the liability for failure to comply with the guarantee. However, section 67 of the ACL does exist, and the courts will endeavour to give it some meaning and scope for operation, so that it operates harmoniously with section 64 and the other provisions of the ACL describing the scope of the consumer guarantees and the remedies for contravention of those guarantees.
There is a respectable argument that the existence of section 67(a) evinces an intention on the part of the legislature that the consumer guarantee regime was never supposed to apply to the supply of goods to Australian consumers where the proper law of the supply contract is the law of a jurisdiction outside Australia. Test it this way. If the consumer guarantee regime was intended to apply to all consumer supply contracts with a sufficient connection to Australia regardless of their proper law, why has the legislature sought to confirm (by enacting section 67(a)) that the consumer guarantee regime applies to a narrower class of contracts, namely those with Australian law as their proper law? In other words, the argument is that the legislature considered that section 67(a) was necessary because the other consumer guarantee provisions of the ACL do not apply if the proper law of the contract is not the law of an Australian jurisdiction; the very existence of section 67(a) is what qualifies the operation of the remainder of the ACL consumer guarantee provisions.
If section 67(a) did not operate in this manner, what function would it play in the scheme of the ACL? The guarantees themselves would apply to all supplies of goods or services to consumers where such supplies had a sufficient connection to Australia, and section 64 would apply to all consumer supply contracts with a sufficient connection to Australia, no matter what their proper law was, so as to render void any contractual terms which were inconsistent with the consumer guarantees. In this scenario, section 67(a) would perform no useful function – there would never be any need to engage its provisions. Courts will always strive to interpret legislation in a manner which enables each legislative provision to operate harmoniously with the other provisions of the statute, and they will be very slow to find that a particular statutory provision exists solely to reinforce the result achieved by other provisions of the statute.
What then of section 67(b)? If the effect of section 67(a) is to qualify the remainder of the consumer guarantee provisions of the ACL as set out above, section 67(b) still has some work to do – it would apply to terms in supply contracts where the proper law is Australian law but for some reason the parties have chosen to substitute the provisions of the law of a different jurisdiction in place of the operation of the ACL consumer guarantees. The respective statutory functions of section 67(b) and section 64 of the ACL remain difficult to reconcile. Both seek to restrict the autonomy of contracting parties to reach an enforceable agreement that would cut across the operation of the consumer guarantee regime. In my view section 67(b) addresses one specific means by which parties may attempt to enter into agreements that would subvert the intended operation of the consumer guarantee regime and, given the generality of section 64, section 67(b) should be confined to that narrow method of “substitution”. The decision in the Laminex case can be distinguished because:
- the scope of section 67 of the TPA was not the ultimate issue in that case – in a sense the decision was akin to a decision in a strike out claim, in that the effect of the decision was merely to allow the case to proceed on its merits (albeit in proceedings to be commenced in another jurisdiction);
- the court did not refer to section 68 of the TPA in its reasoning; and
- the role of section 67 of the ACL is quite different from the role played by section 67 of the TPA because of the different framework for remedies for breach of the ACL consumer guarantees.
Speaking of the interaction between statutes such as the TPA and the ACL and contracts governed by the law of a foreign country, the authors of Australia’s leading text book on conflicts of laws, Nygh’s Conflicts of Laws state (at pars [19.47 and 19.48] of the latest edition):
“It was never authoritatively determined whether or not the [TPA] could be invoked or would be construed as applying in circumstances where relief was sought under s87 of that Act in relation to a contract governed by a law other than Australian law. That remains the case under the [ACL] and its various state analogues modelled on the [TPA]. The better view is that such legislation does not apply to contracts which, objectively considered, are governed by the law of a foreign country, or, in the case of state legislation, by the law of another Australian jurisdiction. …
The use of the familiar statutory formulation in s67 of the [TPA] and the [ACL], confined as it is to ‘consumer contracts’, may suggest that the other provisions of the Act such as those providing remedies for contravention of the misleading and deceptive conduct provisions were not intended to be mandatory in the private international law sense of overriding an otherwise legitimately chosen proper law of the contract (at least in circumstances where Australian law would not otherwise have governed the contract). The competing views as to the relationship and interaction between the [TPA], its successor and analogues, on the one hand, and a contract governed by a foreign proper law, on the other hand, are finely balanced, and await authoritative determination at an intermediate or the ultimate appellate level”
Now, it is true that the authors of Nygh’s Conflicts of Laws did not discuss the different roles of s67(a) and s67(b) in the context of contracts to which the consumer guarantee provisions may apply. Nonetheless, it is apparent that the application of the consumer guarantee provisions of the ACL to cross-border internet transactions with Australian consumers is an issue requiring judicial clarification.
Application to Valve’s case
In Valve’s case, the relevant contracts are governed by the law of the State of Washington, USA. Valve may well seek to argue that, putting to one side the express choice of law, the proper law of the contracts was the law of the State of Washington. Factors in support of that argument include the following:
- the place of performance of the contractual obligation to supply was the location of the server from which the software was downloaded by Australian consumers (presumably a location in the USA);
- the currency for payment was USD;
- the place of origin of the software was the USA
It is also possible, depending on the method by which consumers become bound to the relevant contractual terms, that the contract was not made in Australia – that will depend upon the location in which acceptance was communicated to the offeror.
The place of payment will also be relevant. Australian courts have held that “[t]he place of payment depends on the place which, in the absence of express provision in the contract, it should be held from all the circumstances the parties should be taken as impliedly intending.”[1] Presumably most payments are typically made by credit card through “card not present” e-commerce transactions. Valve would, most likely, supply the software on receipt of an authorisation from the card scheme operator (Visa, Mastercard etc). Interestingly, the Visa International Operating Rules provide that the “Transaction Country” is “[t]he Country where a Merchant Outlet is located, regardless of the Cardholder’s location when a Transaction occurs.” It may well be that the place of payment for Valve’s transactions with Australian consumers is somewhere in the USA.
If Valve could make out that the system of law with the closest connection to the transactions was the law of the State of Washington, the opportunity would arise to test the scope of sections 64 and 67 of the ACL and argue that the ACL consumer guarantees did not apply at all. If that argument succeeded, the ACCC’s claims based on misleading and deceptive conduct would fail.
Unsurprisingly, the ACCC’s pleadings do not attempt to deal with these arguments – they simply refer to section 64 of the ACL and assume that section 64 applies even though the governing law of the contract is the law of the State of Washington.
Valve has not entered an appearance in the proceedings yet. That is probably because they are waiting to be formally served – the ACCC will likely need to obtain leave from the Federal Court of Australia to serve the originating process outside Australia (see Federal Court Rules, rule 10.43). But the Federal Court’s jurisdiction is reasonably clear and both Australia and the USA are parties to the Hague Convention on the Service Abroad of Judicial and Extrajudicial Documents in Civil or Commercial Matters, so it is simply a matter of time before service will be effected on Valve. In light of ACCC v Chen and subsequent decisions, it seems unlikely that the Federal Court would uphold an argument by Valve that the court does not have, or ought not exercise, jurisdiction in the matter. The more prudent course, once service has been properly effected, would be to file a defence arguing that the consumer guarantees do not apply because the proper law of the contract is the law of the State of Washington and neither sections 67 nor 64 operate so as to exclude or modify those terms of the contract that are inconsistent with the consumer guarantees.
In theory, there is no reason why Valve and the ACCC would be unable to agree on the underlying facts necessary for the court to determine the proper law of the contracts. If this could be done, the court could decide as a separate question what the proper law of the contracts was and, if the proper law was not the law of an Australian jurisdiction, whether the consumer guarantees (and associated remedies for contravention) applied.
At the very least, if Valve was to pursue this strategy it may maximise their prospects of reaching a settlement with the ACCC. It has been reported that Valve has stated that “[w]e are making every effort to cooperate with the Australian officials on this matter” and that their position is that they do not offer refunds unless required by local law. Any settlement would involve the parties reaching an agreement as to the circumstances in which Australian law requires a refund. However, for now, the ACCC is talking tough so a settlement may not eventuate any time soon.
We will be watching developments with interest.
[1] See, particularly, Mendelson-Zeller Co Inc v T & C Providores Pty Ltd [1981] 1 NSWLR 366.
[2] Earthworks & Quarries Ltd v. FT Eastment & Sons Pty Ltd [1966] V.R. 24, at 26; followed in BP Australia Ltd v Wales [1982] Qd R 386 at 392
Photo Credit: Title: [New York Central Twentieth Century Limited steam locomotive 5453], Creator: Richie, Robert Yarnall, Date: May 12, 1938
