This is a guest post from Dr. Szilárd Gáspár-Szilágyi, Birmingham University and Dr. Maxim Usynin, Copenhagen University
- The Context
Those following the interaction between International Investment Law and EU Law have noticed that the past couple of years have produced a growing number of cases in which the Court of Justice of the EU takes a hostile attitude towards investment treaty arbitration (ITA). In its 2018 Achmea judgment (here and for commentaries here and here) the Court of Justice concluded that investor-state arbitration clauses found in international agreements concluded between EU Member States, ’such as’ the ones in the Netherlands-Slovakia BIT, were precluded by Articles 344 and 267 TFEU (we use this very specific language from the original judgment, as the Court in European Commission v. Micula phrases its holdings in Achmea more broadly, see para. 138, [here]).
Things have then picked up in late 2021, the Court delivering its judgments in Komstroy (here) and PL Holdings (here). In Komstroy, the Court of Justice held obiter dicta that the Energy Charter Treaty was also incompatible with EU law in an intra-EU setting, even though the referring national court did not raise any questions concerning compatibility and the original arbitration was not an intra-EU arbitration (for commentaries here). Then, in PL Holdings, the Court held that national legislation, which would permit the circumvention of Achmea, by allowing a Member State to conclude with an investor an ad hoc arbitration agreement that is identical in terms to an arbitration clause in an intra-EU BIT, is also precluded by Articles 267 and 344 TFEU.
This was then followed by the 25th of January 2022 judgment in the European Commission v. Micula et al appeal (here), in which the Court of Justice set aside the judgment of the General Court of the EU and referred the case back to the lower court to adjudicate on several pleas and arguments. This much awaited judgment is part of a seemingly never-ending string of enforcement cases that followed the 2013 ICSID Award in the Micula v. Romania cases [here].
In short, Romania, prior to its accession to the EU in 2007 had given certain tax incentives for a period of 10 years to two Swedish citizen investors, who invested in an economically deprived area of the country. Before its accession to the EU, Romania had to bring its laws in line with EU law and decided to revoke the tax advantages given to the two investors, for those advantages would be considered illegal state aid under EU law. The investors commenced an ICSID arbitration in 2005 and in 2013 the arbitral tribunal held that Romania had breached the Fair and Equitable Treatment (FET) standard of the Sweden-Romania BIT of 2002. Romania then began to repay part of the award, but the EU Commission issued Decision 2015/1470 in which it held that repayment of the damages awarded by the arbitral tribunal constitute illegal state aid under EU law. This Decision was challenged by the investors under Article 263 TFEU (annulment proceedings) before the General Court of the EU, which annulled the Decision [here]. This led to the appeal that we currently analyse.
This of course is not the end of the story, because in the meantime the investors tried enforcing the ICSID award in the USA [here], Belgium [here], Sweden [here], the UK [here], and Romania [here], some of the cases having been won by them and others lost. These cases raise a myriad of legal issues, but in this short blogpost we will focus only on the recent judgment of the Court of Justice in European Commission v Micula, and even within that judgment we are only interested in para. 144 (below). This paragraph leads us to believe that the Court of Justice’s antagonistic attitude towards investment treaty arbitration might also be because the Court misunderstands the role of investment law and the objectives of investment treaty arbitration.
- The Problematic Statements
The Court of Justice in essence held that the General Court erred in its judgment when it concluced that the EU Commission had no competence to decide on state aid. The Court then relied on Achmea, Komstroy and PL Holdings to argue that Romania’s consent to arbitrate ’lacks any force’ from the moment it entered the EU (para. 145). According to the problematic paragraph:
„Such consent [under a BIT], unlike that which would have been given in commercial arbitration proceedings, does not originate in a specific agreement reflecting the freely expressed wishes of the parties concerned, but derives from a treaty concluded between two States in the context of which they have, generally and in advance, agreed to exclude from the jurisdiction of their own courts disputes which may concern the interpretation or application of EU law in favour of arbitration proceedings.” [emphasis added]
We believe that this paragraph contains two misunderstandings about how investor-state arbitration and BITs work; the first concerns the nature of consent to arbitrate under an investment agreement and the second the purpose of investor-state dispute settlement (ISDS).
2.a. The Nature of Consent
Consent to arbitrate under an investment treaty is a long-standing issue in investment treaty arbitration and it goes back to the initial cases, such as SPP v. Egypt for domestic investment laws and AAPL v. Sri Lanka for investment treaties, which established that the arbitration agreement is formed in a more peculiar way compared to commercial arbitration. Masterfully described by Jan Paulsson in his ‘Arbitration without privity’ article [here], this development became a ‘silent revolution’ [here] of the then-nascent legal field. After nearly three decades, this revolution is now accepted as standard practice. According to it, the State’s consent in the treaty to submit the dispute to arbitration is a standing offer to arbitrate, which is accepted by the investor when they decide to bring an arbitral claim against the State. When the Court of Justice argues that this type of consent “does not originate in a specific agreement reflecting the freely expressed wishes of the parties concerned”, the Court basically questions the standard practice under investment treaty arbitration. However, is the Court of Justice right in doing so?
Firstly, to say that the wishes of the parties concerned were not ‘freely’ expressed is simply factually incorrect. States willingly sign investment treaties for various reasons, but they are not coerced to do so. The treaties are only void if state consent was coerced by the threat or use of force (Article 52 VCLT), which is ‘generally held’ to refer solely to armed force [here]. An investor also willingly brings arbitral claims against the host State. Therefore, both parties to the dispute have without a doubt freely expressed their wishes.
Secondly, what does the court mean by ‘specific agreement’? Would it have to be a separate, written arbitral agreement, as opposed to the accepted practice of a standing offer made by the host State in the treaty which is accepted by the investor when it brings a claim? If so, many commercial arbitration agreements would also not meet this standard. Some are included as arbitration clauses at the end of contracts (or as a reference to some general terms and conditions), others are concluded separately as compromis after the dispute has arisen, while others can be included in an exchange of emails or other forms of electronic messages. At the end of the day, an agreement is the acceptance by one party of an offer made by another. If this is the standard definition of an ‘agreement’, then the standard practice in ITA meets this definition. In other words, the Court of Justice seems to (whether wilfully or not) mischaracterize the nature and even the existence of an agreement to arbitrate if the offer is contained in a treaty and the offer is accepted when the claimant initiates arbitration. To take it one step further, what about those situations when ISDS provisions in treaties provide for ‘arbitration under any other rules of arbitration agreed upon by the parties’ and not just ‘standard’ venues such as ICSID or ad hoc UNCITRAL arbitration? If the parties (the investor and the host State) can agree to any arbitral rules, then they also have to agree to arbitrate.
Thirdly, playing devil’s advocate, if we agree that consent in ITA is its Achilles’ heel, because it is based on an accepted legal construct and it is not expressly stated in investment treaties, then the same is true for the ‘autonomy of EU law’. This nebulous term is used by the Court to substantiate its recent decisions against ITA in an intra-EU setting [here], but at the root of it, it rests on the Court’s holding in the seminal Van Gend en Loos case that the EU Treaties have created a ‘new legal order of international law’ [here], a statement which to this day has not been included in the EU Treaties.
2.b. The Purpose of ISDS and ITA
The second issue is a clear case of the Court of Justice mischaracterising or misunderstanding the purpose of investment treaties and ITA. According to the Court, EU Member States had “agreed to exclude from the jurisdiction of their own courts disputes […] in favour of arbitration proceedings”. For the sake of this argument, it is not relevant whether EU law would be interpreted by a non-EU tribunal (one of the main concerns of the Court in such cases; yet, it should be noted, in passing, that the claimed monopoly of interpretation threatens the Court of Justice’s coexistence with virtually any other international court) [here].
This passage indicates that the Court does not fully understand ISDS and the purpose of BITs. BITs are meant to promote investments (there is a long-standing discussion whether they actually do) and to protect them. Protection in turn is ensured by the inclusion of international standards of protection (national treatment, MFN treatment, FET, etc) and the presence of ISDS. However, ISDS provisions are not uniform, and they are not restricted to ITA. Some investment treaties do not provide for international arbitration, others provide for a choice between domestic courts, alternative methods such as conciliation and mediation, and/or treaty-based arbitration. Then there are also fork-in-the road and no-U turns clauses [here].
Whilst investment treaty arbitration has received most of the criticisms and legitimacy challenges in the last decade or so, as we can see, ISDS provisions in investment treaties are not confined to ITA and include several other venues, including domestic courts, which as one of us has shown [here], foreign investors do use even when ITA is a possibility. Furthermore, nowhere do BITs mention that their objective is to remove cases from domestic courts. In other words, the Court’s argument clearly shows a misunderstanding of how BITs and ISDS work, and how ITA is only one way of settling investor-State disputes under a treaty.
- Why this is important
Why should all of this matter, one might ask. It should matter because it shows that when a court which is foreign to a system and uses the features of that system to obtain its own goals, the chances that the foreign system will be misunderstood and mischaracterised can be very high. In this case the Court of Justice’s animosity towards ITA has resulted in protracted enforcement cases, disgruntled investors that cannot enforce their awards in the EU, EU Member States which do not know whether they should abide by their international or EU obligations, foreign courts which do not understand EU law, and arbitral tribunals which have become increasingly hostile towards EU law [here]. Sensing increasing legal instability, Lavranos expects a broader avoidance of the EU as a seat of arbitration or enforcement, which will likely favour other venues with more arbitration-friendly regimes [here]. Moreover, there are the non-legal issues, such as the negative message the EU sends to foreign investors, corporate restructuring to take advantage of non-intra-EU investment treaties, and the possibility that some EU States when they get into financial difficulties will not be able to raise capital on the international markets or from the IMF, until their ICSID debts are not settled, as the case of Argentina illustrates [here]. Lastly, the argument which ties compliance with international legal obligations, as manifested in enforceable arbitration awards, with the social identity of democratic states deserves merit in EU cases as well [here].
