Choice of law rules and statutory interpretation in the Ruby Princess Case in Australia


 Written by Seung Chan Rhee and Alan Zheng

Suppose a company sells tickets for cruises to/from Australia. The passengers hail from Australia, and other countries. The contracts contain an exclusive foreign jurisdiction clause nominating a non-Australian jurisdiction. The company is incorporated in Bermuda. Cruises are only temporarily in Australian territorial waters.

A cruise goes wrong. Passengers, Australian and non-Australian, want relief under the Australian Consumer Law (ACL). They commence representative proceedings alleging breaches of consumer law, and negligence in the Federal Court of Australia. The Australian court must first resolve the conflict of laws problems posed – problems as sustained as they have been complex in the history of private international law.

These are the facts at the heart of the Ruby Princess cruise, and her 2,600 passengers. The story was reported widely. A COVID-19 outbreak prematurely terminated the cruise. Many passengers contracted COVID-19; some died. Unsurprisingly, the cruise then spawned an inquiry and a class action against Carnival plc (Carnival) as charterer and operator of the Ruby Princess, and Princess Cruise Lines Ltd, the Bermuda-registered subsidiary and vessel owner.

Statute has left little of the common law untouched. This short note analyses the interaction between a mandatory law and an exclusive jurisdiction clause in the context of the case. The note observes the tension between the selection of the statutist approach or conventional choice of law rules as an analytical starting point, in difficult consumer protection cases.


The Ruby Princess’ passengers contracted on different sets of terms and conditions (US, UK and AU). The US and UK terms and conditions contained exclusive foreign jurisdiction clauses favouring the US and English courts respectively (PJ, [26], [29]). US customers also waived their rights to litigate in representative proceedings against Carnival (the ‘class action waiver’) (PJ, [27]). In aid of these clauses, Carnival sought a stay of the proceedings vis-à-vis the UK and US passenger subgroups.

Whether a stay is granted under Australian law turns on whether the Australian court is ‘a clearly inappropriate forum’ (See Oceanic Sun Line Special Shipping Co Inc v Fay at 247–8) (Oceanic Sun Line). In Regie Nationale des Usines Renault SA v Zhang (Renault v Zhang), the High Court (at [78]) described the test as requiring the applicant to show the Australian proceeding:

would be productive of injustice, because it would be oppressive in the sense of seriously and unfairly burdensome, prejudicial or damaging, or vexatious …

In Voth v Manildra Flour Mills Pty Ltd (Voth), a majority observed (at 566):

the extent to which the law of the forum is applicable in resolving the rights and liabilities of the parties is a material consideration … the selected forum should not be seen as an inappropriate forum if it is fairly arguable that the substantive law of the forum is applicable in the determination of the rights and liabilities of the parties.

Through these cases the High Court elected not to follow the English approach (see Spiliada Maritime Corporation v Cansulex Ltd) which requires that another forum is clearly or distinctly more appropriate. The Australian test, after Voth poses a negative test and a more difficult bar.

First Instance

Stewart J found the Federal Court was not a clearly inappropriate forum and declined to stay the proceedings. A critical plank of this conclusion was the finding that the exclusive foreign jurisdiction and class action waiver clauses were not incorporated into the contracts (PJ, [74]). Even if the clauses were incorporated, Stewart J reasoned in obiter that the class action waiver was void as an unfair contract term under s 23 of the ACL (PJ, [145]) and the Federal Court was not a clearly inappropriate forum.

As noted in Voth and Oceanic Sun Line, simply because the contract selected the US or UK as the particular lex causae did not end the analysis (PJ, [207]) — the US and UK subgroups were not guaranteed to take the benefit of the ACL in the US and English courts, notwithstanding Carnival’s undertaking that it would not oppose the passengers’ application to rely on the ACL in overseas forums (PJ, [297], [363]). Ultimately, there remained a real juridical advantage for the passengers to pursue representative proceedings together in Australia.

Carnival appealed.

Full Court

The majority (Derrington J, Allsop CJ agreeing) allowed Carnival’s appeal, staying the US subgroup’s proceedings. Unlike the primary judge, the majority reasoned the clauses were incorporated into the US subgroup contracts. Further, a stay should be refused because the US and English courts had similar legislative analogues to the ACL (FCAFC, [383]-[387]). Although he US passengers would lose the benefit of the class action, that was a mere procedural advantage and the question of forum is informed by questions of substantive rights (FCAFC, [388]).

Rares J dissented, upholding the primary judge’s refusal of a stay (FCAFC, [96]).

The passengers appealed to the High Court.

The Interaction between a Mandatory Law and an Exclusive Jurisdiction Clause

Statutes generally fall into one of three categories (see Maria Hook, ‘The “Statutist Trap” and Subject-Matter Jurisdiction’ (2017) 13(2) Journal of Private International Law 435). The categories move in degrees of deference towards choice of law rules. First, a statute may impose a choice of law rule directing the application of the lex fori where a connecting factor is established. Second, a statute may contain, on its proper construction, a ‘self-limiting’ provision triggered if the applicable law is the lex fori. Third, a statute may override a specified lex causae as a mandatory law of the forum. An oft-repeated refrain is that all local Australian statutes are mandatory in nature ([2023] HCATrans 99).

In the High Court, Carnival contended that if contracting parties select a lex causae other than the forum law, the forum statute will not apply unless Parliament has expressly overridden the lex causae.

The passengers (supported by the Commonwealth Attorney-General and ACCC, as interveners) took a different starting point — the threshold question is whether the forum law, as a matter of interpretation, applies to the contract irrespective of the parties’ usage of an exclusive jurisdiction clause. In this case, several factors supported the ACL’s application including s 5(1)(g) of the CCA, and the need to preserve the ACL’s consumer protection purpose by preventing evasion through the insertion of choice of law clauses.

The parties adopted unsurprising positions. The passengers’ case was conventionally fortified by the statutist approach, prioritising interpretation in determining the forum statute’s scope of application. Carnival relied on the orthodox approach, prioritising choice of law rules in controlling when and to what extent forum statutes will apply, and more aligned with comity norms and party autonomy the selection of the governing law of private agreements. The orthodox approach was exemplified in Carnival’s submission that ‘[i]t was not the legislature’s purpose to appoint Australian courts as the global arbiter … of class actions concerning consumer contracts across the world’ (See Respondent’s Outline of Oral Argument, p. 3).

Against that view, it was said that party autonomy should be de-emphasised where contracts are not fully negotiated, involve unequal bargaining power and standard terms (contracts of ‘adhesion’ as here provide a good example): see [2023] HCATrans 99 and the exchange between Gordon J and J Gleeson SC.

As scholars have noted, differences between the two approaches can be almost imperceptible. Characterisation is a ‘species of interpretation’ (Michael Douglas, ‘Does Choice of Law Matter?’ (2021) 28 Australian International Law Journal 1). However, the approach taken can lead to different outcomes in hard cases.

The key obstacle to the statutist approach is uncertainty. If interpretation of a statute’s extraterritorial scope controls the choice of law, then how do contracting parties ensure their selection of law prevails and that they are complying?

Interpretation (both in the choice of law sense and statutory interpretation) invites reasonable arguments that cut in both directions requiring judicial adjudication. Take, for example, Carnival’s response to the passengers’ argument that the ACL’s consumer protection policy weighs against the use of choice of law clauses to evade liability. Carnival contended any evasion can be controlled by a two-step approach: firstly, applying the ACL’s unfair contract provisions to the choice of law clause itself and, if it the clause is void, only then secondly applying the provisions to the contract as a whole. However, this only shifts the application of statutory interpretation to an anterior stage, namely how and when a given choice of law clause, on its face, might be considered unfair. To the extent any determination of unfairness could be made, this turns on the consequences of the clause per se than any particular manner of wording. Such an outcome equally produces unpredictability as to the anticipated effect and application of the forum law.

There is another example on point. Section 5(1)(g) extends the ACL to the ‘engaging in conduct outside Australia’ by bodies corporate carrying on business in Australia. Carnival’s expressio unius-style argument that s 5(1)(g) does not support the passengers’ case because the unfair contracts prohibition is not predicated on ‘engaging in’ any conduct, whereas ACL prohibitions apply to ‘conduct’. Accordingly, taking up a point made by the Full Court majority (FCAFC, [301]), Carnival contended a limitation should be read into s 5(1)(g) else it capriciously apply to companies like Carnival whose business were entirely engaged outside of Australia’s territorial limits.

Nevertheless, as the appellants pointed out (relying on drafting history), ‘when the unfair contract terms legislation was first introduced … s 5(1) was specifically amended to apply to those provisions’ (See Appellant’s Written Submissions, p. 6).  It is therefore apparent how the statutist approach invites a certain level of textual skirmishing.

Choices are available to judges under both the statutist approach and in the application of choice of law rules (see Michael Douglas, ‘Choice of Law in the Age of Statutes’ in Michael Douglas, Vivienne Bath, Mary Keyes and Andrew Dickinson, Commercial Issues in Private International Law: A Common Law Perspective (Hart Publishing, 2019) ch 9). However, it does not follow that there are comparable levels of certainty in the two approaches. Characterisation of a case as contract or tort (to take a very general example) invites a narrower range of choices than the entire arsenal of statutory interpretation techniques deployable analysing words in a statutory provision. That is so because characterisation is controlled by matters external to submissions, namely pleadings and the facts as objectively found (e.g. where was the defective product manufactured, or where was the injury sustained). Interpretation, particularly through the modern focus on text, context and purpose, is not disciplined by facts or pleadings. Instead, it is shaped by submissions and argumentation actuated by the connotative ambiguity found in statute.

That has led the High Court to observe that choice of law rules uphold certainty. In Renault v Zhang, Gleeson CJ, Gaudron, McHugh, Gummow and Hayne JJ stated ([66]-[67]):

The selection of the lex loci delicti as the source of substantive law meets one of the objectives of any choice of law rule, the promotion of certainty in the law.  Uncertainty as to the choice of the lex causae engenders doubt as to liability and impedes settlement.

Against the aim of certainty (and deference to choice of law clauses) are the countervailing considerations arising from legislative policy and the higher-order status of statute over choice of law rules sourced from the common law (see Douglas, ‘Choice of law in the Age of Statutes’). The interveners put it as an ‘unattractive prospect’ if the ‘beneficial’ aspects of the ACL regime could be defeated by expedient foreign jurisdiction clauses.

Insofar as the legislature evinces an intent to confer the benefit of legislation beyond Australia’s territorial bounds, courts bound by an interpretive obligation to give effect to that legislative intention will not be able to defer to choice of law rules. In the case of the CCA and the ACL, s 15AA of the Acts Interpretation Act 1901 (Cth) enjoins courts to prefer the interpretation ‘that would best achieve the purpose or object of the Act (whether or not that purpose or object is expressly stated in the Act)’. Douglas and Loadsman (see ‘The Impact of the Hague Principles on Choice of Law in International Commercial Contracts’ (2018) 19(1) Melbourne Journal of International Law 1) observe that:

It is consistent with this purposive approach to statutory interpretation that Australian courts take a broad approach to the geographical scope of Australian statutes. In an environment where Australian lives and businesses increasingly cross borders on a regular basis, it would defeat the purposes of many pieces of Australian legislation if courts were to take a territorially-limited approach to statutes’ scope of operation.

No doubt there is some truth to Carnival’s submission that Parliament did not intend to render Australian courts the global arbiters of consumer contracts. However, subject to a pronouncement to the contrary from the High Court, the judgments to date in Karpik v Carnival plc suggest a statutist analysis, however uncertain, difficult or comity-ablating, will be a necessary precondition to determining the weight given to the wording of a choice of law clause. This is ultimately a consequence of the premium placed on a purposive construction to mandatory laws arising out of the home forum. For better or worse (and a strong case has been made for worse – see Maria Hook, ‘The “Statutist Trap” and Subject-Matter Jurisdiction’ (2017) 13(2) Journal of Private International Law 435), ‘[i]f the purposive approach to statutory interpretation gives rise to forum shopping in favour of Australian courts, so be it’ (see Douglas and Loadsman, 20).

Notwithstanding this, another difficulty with Carnival’s submissions in favour of the choice of law approach is that it functionally revives the common law presumption of non-extraterritorial application of laws and elevates the rebuttability threshold of that presumption to something made ‘manifest’ by parliament (which has been keenly disputed in the High Court: see Respondent’s Submissions, [10]).

It is important to recall that the presumption was always couched in the language of construction. In Wanganui-Rangitiei Electric Power Board v Australian Mutual Provident Society, Dixon J stated (at 601):

The rule is one of construction only, and it may have little or no place where some other restriction is supplied by context or subject matter.

Rebuttability does not arise at all if the context or subject matter of the forum statute, as a matter of interpretation, supplies a relevant territorial connection. If it so supplies, that territorial connection operates as a restriction.

Dixon J also went on to state (at 601):

But, in the absence of any countervailing consideration, the principle is that general words should not be understood as extending to cases which, according to the rules of private international law administered in our courts, are governed by foreign law.

Most recently in BHP Group Ltd v Impiombato, Kiefel CJ and Gageler J (at [23]) considered the common law presumption resembled a ‘presumption in favour of international comity’ rather than one against extraterritorial operation – although it is worth noting that three other judges recognised (at [71]) the common law presumption was ultimately a statutory construction rule which did not always require reference to comity. Nevertheless, an important factor for Kiefel CJ and Gageler J in finding the class action provisions of Part IVA of the Federal Court of Australia Act 1976 (Cth) were not restricted to Australian residents by the presumption was the fact no principle of international law or comity would be infringed by a non-consenting and non-resident group member being bound by a judgment of the Federal Court in relation to a matter over which that court had jurisdiction.

Conversely, as Derrington J noted on appeal (FCAFC, [300]), the extension of s 23 to the transactions of companies operating in overseas markets as a result of their ancillary dealings in Australia would have been an ‘anomalous result’. Such a result would not have promoted comity between Australia and other national bodies politic, where the ACL would have had the result of potentially subjecting foreign companies to obligations additional to those imposed by the laws of their home country. As Carnival put it in the High Court:

if a company happens to carry on business in Australia, all of its contracts with consumers (as defined) all over the world are then subject to Part 2-3 of the ACL. It would mean, for example, that contractual terms between a foreign corporation and consumers in Romania under standard form contracts can be deemed void under s 23 (Respondent’s Submissions, [36]).

Without an expressed intention to the contrary, it was unlikely that Parliament had intended to ‘legislate beyond the bounds of international comity’ – into an area that would ordinarily be expected to be governed by foreign law.

To some extent, the judgments to date, despite their differing conclusions, suggest in common that an entirely non-statutist outcome (insofar as the CCA and ACL is concerned) is something of a will-o’-the-wisp. If it is accepted that matters of high forum public policy can supervene the contractual arrangements of the parties, expressed in no uncertain terms, then a court must always evaluate legislation in a statutist manner to determine how contractual arrangements interact with that policy. This is so even if, as in Derrington J’s view in Carnival plc v Karpik, the conclusion would be that the policy would not be advanced by applying the mandatory law.

The High Court’s decision will not only clarify the ambit of the CCA regime; it will materially bear upon the desirability of Australian courts as a forum for future transnational consumer law class actions. Coxtensively, companies with Australian operations liable to be on the respondent end of such class actions will be watching the developments closely before drafting further exclusive foreign jurisdiction clauses.

Judgment is reserved in the High Court.

Seung Chan Rhee is a solicitor at Herbert Smith Freehills. Alan Zheng is an Australian-qualified lawyer at Linklaters LLP. The views in this note are the views of the authors alone. The usual disclaimers apply.