Investor-State Dispute Settlement (ISDS) clauses are a prominent feature of many modern International Investment Agreements (IIAs). They are included in nearly all the IIAs to which Australia is a party. Typically, an ISDS clause allows a foreign investor (often a corporation) to challenge a government decision before a panel of private arbitrators who have the power to make decisions and make awards that are binding and enforceable.

Despite their prevalence, ISDS clauses, and the machinery in place for enforcing them, are controversial. Some argue that these clauses undesirably constrain national sovereignty, by subjecting national governments to the decisions of international arbitral bodies. This might be of some benefit when the nation in question has inadequate rule of law protections, but is problematic when the country in question has a well-functioning legal system with a robust judiciary, and when national legislation may be pursuing policy measures that serve important social purposes. The arbitral bodies established to adjudicate ISDS disputes have also been criticised, for lacking independence, transparency, and fair procedures. But do any of these concerns translate into problems of constitutional validity? That is, is there any reason to suspect that national governments lack the constitutional capacity to enter into agreements that contain ISDS provisions?

As recent events (from the UK government’s thwarted attempt to trigger Brexit without parliamentary authorisation (see here and here), to South Africa’s stalled attempt to exit the ICC) remind us, every government’s capacity to engage with the international legal system is constrained — to varying extents — by their domestic constitutional arrangements. The intersection between international and domestic constitutional law raises many important questions of law and politics. In this post, we consider whether there are any principles of Australian constitutional law that might constrain the Australian government’s capacity to enter into agreements containing ISDS clauses.

Like many other countries, Australia is a dualist system. The Australian Constitution s 61 gives the executive government the power to enter into international agreements on Australia’s behalf. However, international agreements do not have domestic legal effect, unless and until they are implemented by legislation validly enacted by the federal Parliament (for statements to this effect, see Dietrich v The Queen (1992) 177 CLR 292; Kioa v West (1985) 159 CLR 550, 570 (Gibbs CJ)).

It is generally assumed that the Australian government can enter into international agreements without parliamentary authorisation. But under the Australian Constitution, the executive power is not unlimited: its scope is informed and constrained by the Constitution. These constraints have potential implications for the government’s power to enter into agreements containing ISDS clauses.

In an important article, the former Chief Justice of Australia’s High Court, Robert French, queried whether ISDS tribunals were effectively “a cut above the courts” that is, whether entry into agreements containing ISDS provisions empowers these tribunals to second guess or undermine the decisions of Australia’s judiciary. The Australian judiciary is established by the Australian Constitution, and robustly independent. The High Court sits at the apex of the judicial system, and a key part of its role is to determine the validity of Australian government and legislative action. This could give rise to tension where actions that have been found to be valid by the High Court are then questioned before an international tribunal. Indeed, the decision of the High Court could itself be the action that triggers an ISDS dispute.

While this may legitimately be cause for concern, we do not think it translates into a problem of constitutional validity. That is, the constitutional principles concerning the independence and authority of the High Court are not infringed by Australia entering into agreements containing ISDS provisions. If an ISDS arbitrator were to find that action taken by Australia is contrary to the relevant agreement, this would not affect its validity as a matter of Australian domestic law. Hence, a decision of an ISDS arbitrator cannot undo a decision of the High Court, whose domestic legal superiority in this regard is maintained. Undoubtedly, the risk of an adverse finding by an ISDS tribunal may have a chilling effect on governmental action, and such a decision may in a practical sense undermine a High Court decision that such action is legally valid – but strictly speaking, the tribunal does not usurp the High Court’s powers.

However, there is another constitutional principle, which we think is a more plausible constraint on the Australian executive’s power to enter into agreements containing ISDS provisions. The application of this principle to ISDS has not yet been noted, not least – we think – because the principle itself is fairly “new”, having only been articulated by the High Court in the case of Williams v Commonwealth [2012] HCA 23 (Williams). In this case, the High Court held that the power of the Australian executive to enter into agreements and spend public moneys was not unlimited, nor as broad as it might be in other jurisdictions. The principles of federalism in conjunction with those of responsible government mandate that, whenever the executive agrees to expend public moneys (other than in the ordinary course of government), parliamentary approval is required. This is not least because the upper house of the Australian Parliament (the Senate), in which all states are represented equally notwithstanding their disparate populations, must be given some say in the expenditure of public moneys.

We think that it is at least plausible to suppose that this principle constrains the executive’s ability to enter agreements that contain ISDS provisions, too. Arguably, an agreement containing an ISDS provision can be characterised as a conditional agreement to pay — for if Australian breaches the obligations under that agreement, it is exposed to ISDS arbitration, the awards of which are automatically enforceable in every jurisdiction which is party to the New York Convention or ICSID Convention, depending on the tribunal that made the award. Thus it might be argued that the executive cannot enter into this type of agreement without parliamentary authorisation.

It seems quite clear that the Australian Parliament would be capable of providing that authorisation – even retrospectively, so as to cover agreements entered into in the past. Nonetheless, this requirement is an important one. It means that the Parliament has some say over which agreements containing ISDS provisions the executive can enter, and which it cannot. If all agreements containing ISDS provisions must be effectively authorised by Parliament, then some of the concerns created by these provisions may be ameliorated – at least to some extent. Specifically, claims that ISDS diminishes sovereignty, or allows the actions of a democratically elected Parliament to be undermined by an arbitral tribunal, would lose some of their bite, because all of this would have been authored by that same Parliament. On the other hand, many aspects of the Williams principle — and its application to ISDS – remain uncertain.


The authors are Lisa Burton Crawford, Patrick Emerton and Emmanuel Laryea. They explore these issues further in “Investor-State Dispute Settlement and the Australian Constitutional Framework”, forthcoming in Colin Picker, Heng Wang, and Weihuan Zhou (eds) The China Australia Free Trade Agreement: A 21st-Century Model (Hart).