Professor Steve Peers, University of Essex
In recent years, investor-state dispute settlement (ISDS) has become a political minefield. Its critics argue that ISDS is a secret court system designed to allow multinational corporations to thwart any progressive legislation approved by democratically elected governments. Its defenders argue that these claims are exaggerated, and that ISDS performs a useful function attracting investment and securing property rights.
The arguments about ISDS are worldwide, but they have increasingly arisen within the particular framework of EU law. From 2009, the Treaty of Lisbon gave the EU exclusive competence over foreign direct investment as part of its common commercial (trade) policy, alongside goods, services and trade-related aspects of intellectual property. EU trade policy developed to include negotiations for ISDS as part of trade negotiations (although pre-existing investment treaties between EU Member States and non-EU countries were grandfathered, with a process in place to regulate negotiation of such treaties in future).
However, this led to political and legal difficulties in negotiating trade agreements: the former because of public concern about ISDS in both the EU and the non-EU countries, and the latter because of uncertainty about whether the EU had sole competence to negotiate treaties with ISDS provisions, or shared it with the Member States. Shared competence means that Member States have to become parties to the treaties concerned, meaning unanimity is required to agree them and there is a process of national ratification, although in practice the EU and the non-EU countries concerned often agree to provisional application of the trade-related parts of the treaty pending such national ratification.
The legal position was clarified when the CJEU ruled in 2017 that ISDS, like 'portfolio' investment (ie non-controlling shares in companies) did not fully form part of the common commercial policy (CCP), but was rather a shared competence between the EU and its Member States. (There’s EU legislation dividing up responsibility between the EU and its Member States in the event of successful investor claims.) Otherwise the Court took a broad view of the scope of the CCP. Coupled with the political concerns about ISDS, this was an opportunity to rethink the role of ISDS in trade policy, either leaving it out of talks completely (Australia and New Zealand, along with the mandate for stripped back trade negotiations with the USA), concluding a trade agreement without insisting on an investment agreement (Japan), or separating the issues into two distinct treaties (Singapore and Vietnam). This revised approach, splitting up trade and investment on a case by case basis, was confirmed more broadly by a Commission communication of 2017 and subsequent Council conclusions in 2018.
In parallel to these developments, the EU responded to concerns about the legitimacy of ISDS by seeking to reform it into a system more palatable to Main Street, rather than Bay Street, as a centre-left Canadian politician might say. A discussion paper of 2015 sums up the Commission’s approach, in particular securing greater transparency, limiting the scope of controversial provisions of investment law, confirming the ‘right to regulate’, and transforming investment tribunals into a quasi-judicial system, with the longer-term intention of establishing a multilateral investment court. Subsequently, the Commission tabled a proposal for such court, and the Council approved a negotiation mandate to that end. (For further details of the negotiations, see here).
Many EU trade and investment policy disputes came to a head early in 2017, when there was a delay in approving the Canada-EU Free Trade Agreement (CETA) because of concerns about ISDS and other issues in one Belgian region. (There were also national constitutional court proceedings challenging CETA in France and Germany, as well as an EU General Court judgment on whether a European Citizens’ Initiative could be launched to stop its ratification). This kerfuffle, coming shortly before the CJEU ruling clarifying the scope of the EU’s common commercial policy, partly prompted the move to downgrade investment objectives in EU trade policy, as discussed above. And part of the overall settlement of the dispute over CETA was the Belgian government asking the CJEU whether the CETA ISDS rules – renegotiated in light of the reformed approach to ISDS – were compatible with EU law.
The Belgian government’s request – submitted on the basis of Article 218 TFEU, which allows the CJEU to rule on proposed international treaties – was answered by the Court of Justice this week (Opinion 1/17). In the meantime, in its judgment in Achmea (discussed here) the CJEU had found that investment treaties between Member States were potentially incompatible with EU law. Those treaties were duly wound up, but it remained to be seen if the Court would have the same concerns about investment treaties with non-EU countries. More generally, the Court has always had concerns about protecting the autonomy of EU law from international courts (see, for instance, Opinion 2/13 on accession to the ECHR, discussed here). Could these concerns about autonomy possibly be reconciled with the nature of ISDS tribunals?
First of all, the CJEU ruled that the case was admissible. Although the Court only has jurisdiction to rule under Article 218 as long as a treaty has not yet entered into force, the provisional application of the trade provisions of CETA did not stand in the way of the Court’s jurisdiction. (Indeed, it appears that the Court would have found the case admissible even if the whole of CETA, including the investment disputes section, was in force provisionally).
The Court then examined the compatibility of the CETA investment provisions with EU law from three angles: the autonomy of the EU legal order; equal treatment and effectiveness; and the right of access to an independent tribunal. In each case, the Court set out the principles and then applied them to CETA.
On the the autonomy of the EU legal order, the Court first recalled its case law that in principle, the EU could sign up to an international treaty which created an international court which could give rulings binding the EU. However, the Court also recalled that any such planned international court cannot infringe the autonomy of EU law. (In practice, the Court has usually been quick to complain that such courts do raise an autonomy problem). This autonomy is, in particular, guaranteed by the EU’s judicial system, which provides for ‘national courts and tribunals and the Court to ensure the full application of that law in all the Member States and to ensure effective judicial protection, the Court having exclusive jurisdiction to give the definitive interpretation of that law’.
For the Court, the crucial factor was that ‘the envisaged ISDS mechanism stands outside the EU judicial system’. The CETA investment court system created by CETA is not part of the domestic court system of Canada, the EU or its Member States. This did not necessarily mean that the ISDS system ‘adversely affects the autonomy of the EU legal order’, because as regards international treaties, the EU judicial system ‘does not take precedence over either the jurisdiction of the courts and tribunals of the non-Member States with which those agreements were concluded or that of the international courts or tribunals that are established by such agreements’. While those treaties form part of EU law and ‘may therefore be the subject of references for a preliminary ruling’ to the CJEU, they ‘concern no less those non-Member States and may therefore also be interpreted by the courts and tribunals of those States’. The ‘reciprocal nature’ of international treaties means that the EU can sign up to treaties creating an international court that is not bound by the interpretations of that treaty given by the courts of any of its parties.
But while EU law did not prevent the creation of such courts, it did place limits on what they could do: ‘they cannot have the power to interpret or apply provisions of EU law other than those of the CETA or to make awards that might have the effect of preventing the EU institutions from operating in accordance with the EU constitutional framework’. It was therefore necessary to address two points: (a) no power for the CETA bodies ‘to interpret or apply EU law other than the power to interpret and apply the provisions of that agreement having regard to the rules and principles of international law applicable between the Parties’; and (b) no power to impact EU law indirectly, by issuing ‘awards which have the effect of preventing the EU institutions from operating in accordance with the EU constitutional framework’.
On the first point, CETA explicitly specifies that its bodies will not have jurisdiction ‘to determine the legality of a measure, alleged to constitute a breach of this Agreement, under the domestic law of a Party’. This was different from treaties which the CJEU had criticised in the past, which would have given an international court the power to interpret EU law. In particular, it was different from an investment treaty between Member States only (which the Court criticised in Achmea), because the EU law ‘principle of mutual trust’…. ‘is not applicable in relations between the Union and a non-Member State’.
Furthermore, the Court was pleased that the CETA investment bodies could not determine the division of powers between the EU and its Member States, unlike the treaty on accession of the EU to the ECHR (on the Court’s ruling in the latter case, see my discussion here). This distinction between the international and domestic systems was consistent with the lack of a prior role for the CJEU, or any power of the CETA bodies to send a reference for a preliminary ruling to the CJEU. It was also consistent with the lack of any national court review of an investment body decision.
On the indirect impact point, several Member States were concerned that a CETA tribunal might rely on the EU Charter ‘freedom to conduct business’ to rule on whether an EU measure is ‘fair and equitable’ under investment law, or ‘whether it constitutes indirect expropriation’, or it is ‘an unjustified restriction on the freedom to make payments and transfers of capital’ as defined in CETA. The CJEU noted that the provisions of CETA were broad and the EU could not block a decision being made against it or an obligation to pay damages, and that a challenger under the CETA investment rules could concern an EU measure ‘of general application’. There was a risk that a series of damages awards might mean that the EU decides to give up the level of protection concerned. Such an indirect impact could, in principle, be incompatible with EU law:
150 If the Union were to enter into an international agreement capable of having the consequence that the Union — or a Member State in the course of implementing EU law — has to amend or withdraw legislation because of an assessment made by a tribunal standing outside the EU judicial system of the level of protection of a public interest established, in accordance with the EU constitutional framework, by the EU institutions, it would have to be concluded that such an agreement undermines the capacity of the Union to operate autonomously within its unique constitutional framework.
In this context, the Court asserted that ‘EU legislation is adopted by the EU legislature following the democratic process defined in the…Treaties’, subject to EU ‘principles of conferral of powers, subsidiarity and proportionality’, and subject to judicial review by the CJEU ‘to ensure review of the compatibility of the level of protection of public interests established by such legislation with, inter alia, the…Treaties, the Charter and the general principles of EU law’.
However, the Court was satisfied that there were enough safeguards against this indirect impact upon EU law. One provision of CETA states that the investment rules:
…cannot be interpreted in such a way as to prevent a Party from adopting and applying measures necessary to protect public security or public morals or to maintain public order or to protect human, animal or plant life or health, subject only to the requirement that such measures are not applied in a manner that would constitute a means of arbitrary or unjustifiable discrimination between the Parties where like conditions prevail, or a disguised restriction on trade between the Parties.
So the CETA Tribunal ‘has no jurisdiction to declare incompatible with the CETA the level of protection of a public interest established by the EU’ in such cases. Therefore it could not ‘order the Union to pay damages’. The Court was also reassured by provisions that state that parties can ‘regulate within their territories to achieve legitimate policy objectives, such as the protection of public health, safety, the environment or public morals, social or consumer protection or the promotion and protection of cultural diversity’, and that regulation which ‘negatively affects an investment or interferes with an investor's expectations, including its expectations of profits, does not amount to a breach of an obligation under this Section’. It also relied upon the Joint Interpretative Instrument to CETA, which states that CETA ‘will … not lower [the standards and regulations of each Party] related to food safety, product safety, consumer protection, health, environment or labour protection’, that ‘imported goods, service suppliers and investors must continue to respect domestic requirements, including rules and regulations’, and that the CETA ‘preserves the ability of the European Union and its Member States and Canada to adopt and apply their own laws and regulations that regulate economic activity in the public interest’.
The Court summed up its view that the CETA bodies’ powers: ‘do not extend to permitting them to call into question the level of protection of public interest determined by the Union following a democratic process’. This was also confirmed by another provision confirming that ‘except in the rare circumstances when the impact of a measure or series of measures is so severe in light of its purpose that it appears manifestly excessive, non-discriminatory measures of a Party that are designed and applied to protect legitimate public welfare objectives, such as health, safety and the environment, do not constitute indirect expropriations’.
While the CETA Tribunal has jurisdiction to apply the broad ‘fair and equitable treatment’ test of investment law, the CJEU was satisfied that this power was limited, only applying to ‘inter alia, situations where there is abusive treatment, manifest arbitrariness and targeted discrimination’. So again, in the Court’s view ‘the required level of protection of a public interest, as established following a democratic process, is not subject to the jurisdiction conferred on the envisaged tribunals to determine whether treatment accorded by a Party to an investor or a covered investment is ‘fair and equitable’.’
More generally, the CETA tribunals ‘have no jurisdiction to call into question the choices democratically made within a Party relating to, inter alia, the level of protection of public order or public safety, the protection of public morals, the protection of health and life of humans and animals, the preservation of food safety, protection of plants and the environment, welfare at work, product safety, consumer protection or, equally, fundamental rights.’ So they did not ‘adversely affect the autonomy of the EU legal order’.
The Court then moved on to the principle of equal treatment and effectiveness. Here, the issue was whether CETA had to be compatible with Article 20 of the Charter (‘equality before the law’) and Article 21(2) of the Charter (non-discrimination on grounds of nationality). On this point, the Court first confirmed long-standing case law that treaties which the EU signed up to had to be compatible with fundamental rights. This issue could also be examined in an Article 218 proceeding, and extended to the Charter. (Indeed, see a 2017 CJEU ruling on another treaty with Canada, concerning the exchange of passenger data, discussed here).
In the Court’s view, Article 21(2) of the Charter did not apply, since it banned discrimination on grounds of nationality only as between EU citizens. However, Article 20 could apply, as its personal scope was not limited. While Article 20 does not oblige the EU to treat all non-EU countries the same (ie, the EU has no internal equivalent to the WTO’s Most Favoured Nation rule), it could apply if there is a difference of treatment within the EU of non-EU citizens on the one hand and EU citizens on the other. As for the principle of effectiveness, it only arose where a CETA Tribunal might find that a fine implementing EU competition law was a breach of the investment guarantees.
Applying these principles, the equal treatment issue was that EU citizens and companies could not invoke the investment provisions in the EU, whereas Canadian citizens and companies could. However, the Court ruled that these two groups were not comparable. The principle of effectiveness was not breached because if the EU or national competition authorities overstepped the limits of EU competition law, their decision could be struck down by the courts anyway.
Finally, as for the right of access to an independent tribunal, the principles were that Article 47 of the Charter bound the EU when entering into international treaties. In the Court’s view, the CETA bodies were very similar to courts, and bound by similar principles of independence. Although the Court was concerned about the accessibility of ISDS for small and medium-sized businesses, it was ultimately satisfied by a statement by the Commission and Council that the issue would be addressed, given that approval of CETA by the EU depended upon that commitment. On the independence of CETA bodies, the Court was satisfied that there was sufficient protection against removal of members, and the rules on payment of members would not preclude their independence. It was unproblematic that the parties could issue a binding interpretation of CETA, since this was a usual feature of international law. In any event, the EU could only agree to interpretations that were compatible with the principles set out in the Court’s opinion, and such interpretations could not have retroactive effects.
First, the Court’s confirmation that the case was admissible is useful. This means that the EU and non-EU countries can decide to apply a treaty provisionally while an Article 218 case is pending before the CJEU. However, this does risk legal complications in the event that the CJEU ultimately finds that the treaty concerned is incompatible with EU law – by analogy with the Council’s statement (no. 20 in the list of statements for the Council minutes) that if a national constitutional court or parliament objects to ratification of CETA, provisional application must be terminated.
As for the substance of the Court’s ruling, its analysis of the equal treatment and effectiveness rules was rather brief. Like the French constitutional court ruling on CETA, there was no clear explanation of why Canadian investors in the EU were in a different position than EU investors. (Possible answers are that the ISDS offers equivalent protection for EU investors in Canada, and that EU investors in the EU can rely on EU internal market law). The assessment of effectiveness takes it for granted that an ISDS body and the EU or Member States’ national courts will reach the same conclusions about the correct application of EU competition law, which is hardly a foregone conclusion. As for the independence of the ISDS system, the Court largely follows its usual approach to defining judicial independence.
The heart of the Court’s judgment is its reconciliation of the autonomy of EU law with the ISDS system. There’s an unusually strong acceptance by the Court of the EU legal system’s co-existence with international law – rather than supremacy over it. But that acceptance is conditional upon the safeguards which the Court then sets out. Here, there is a fundamental tension between the procedural aspect of the ruling (separate court system) and the substantive aspect of preserving the ‘right to regulate’. What if an ISDS body does issue a ruling that arguably infringes the capacity of the EU to decide on the appropriate level of regulation? Given that it’s essential that the ISDS system stands outside the national and EU court systems, how can the boundaries – also essential – which the Court insists must be set upon that system be enforced? The division between ISDS and national courts systems is simultaneously part of the solution and part of the problem.
In short, in British English, the key question for the Court was whether it was willing to throw a spanner into the works of the international investment system. The Court’s answer, in Canadian English, is like having a black fly in your chardonnay.
Is there a way to square this circle? The power of the CETA Joint Committee to issue interpretative rulings would arguably not go far enough to ‘fix’ the problem of an ISDS body ‘running wild’, as such rulings cannot be retroactive and Canada might not agree to them anyway. So let’s return to the courts. The Court rules out a national court review of an ISDS decision. However, it also refers to the possibility of national courts asking the CJEU questions about CETA. Arguably, then, it’s possible to enforce the limits on ISDS bodies by a Member State or the EU refusing to pay a damages award ordered by an ISDS body, leading to a court challenge of that refusal to pay by the winning party – which is technically not a court review of the ISDS body’s decision as such. It would be similar to the well-known case of Kadi, in which the CJEU did not rule on the validity of a UN Security Council measure as such, but on the legality of its application in the EU legal order.
Is the judgment relevant to Brexit? At first sight the judgment is encouraging for those who would like to avoid any role for the CJEU as regards the UK after Brexit, given the Court’s willingness to reconcile the EU legal order with international law. However, that was not the sole factor in the Court’s reasoning, which distinguishes (rather than overturns) prior case law on the autonomy of EU law. A key part of the Court’s reasoning is that the ISDS body, unlike previous international courts which the Court objected to, does not have power to interpret EU law. The position is quite different under the Brexit withdrawal agreement (as I discuss here), and it remains to be seen if it might also be different as regards EU/UK future relationship treaties.
Finally, given that the new ruling concerns a reformed ISDS, how can it be enforced as regards unreformed bilateral investment treaties between EU Member States and non-EU countries (see the most recent list of such treaties here), to the extent that they do not comply with the standards set out by the Court and may apply to issues falling within the scope of EU law? Here the 2012 Regulation grandfathering pre-existing treaties, which also puts in place a process to regulate negotiation of such treaties in future, may be relevant. The review of pre-existing treaties, and control of future treaties, which that Regulation provides for may be applied taking account of the criteria in the Court’s judgment, so as to coordinate updating such treaties to ensure that they are compatible with EU law. This could be similar to the earlier process of updating bilateral aviation treaties between EU Member States and non-EU countries, in light of a series of CJEU judgments on their EU law compatibility.
It's too soon to say whether the reforms of the ISDS, as endorsed by the CJEU in its ruling, will satisfy a sufficient number of critics of the system to reduce the political opposition which ISDS has attracted in the past. Maybe the Court's judgment will turn out to be a death row pardon, two minutes too late. But it's striking that unlike many prior rulings, the CJEU does not appear intrinsically hostile to an international court, but willing in principle to find a way to accommodate it. Furthermore, the constraints the Court insists upon are not justified (as is usually the case) in terms of the Court's own institutional interests in the autonomy of EU law, but in terms of the EU's political institutions' accountability to the democratic process. To adapt the Canadian term, this is a judgment for Main Street, rather than the Kirchberg plateau.
Barnard & Peers: chapter 24
Photo credit: cbc.ca