The enforcement of intra-EU ICSID awards in Germany: anti-arbitration activism by the Bundesgerichtshof?

Luís Heleno Terrinha

Assistant Professor, NOVA School of Law

Of Counsel, PLMJ Lawyers[1]

On 27 July 2023, the Bundesgerichtshof (BGH) handed down a judgement in which it issued an anti-arbitration injunction at the request of the German Federal Republic.[2] The background to this is that Mainstream Renewable Power Ltd. had initiated ICSID proceedings against Germany invoking the latter’s breach of the ECT and requesting compensation of nearly 275 million euros for damages caused by the State’s regulatory overhaul of its activities.[3] The proceedings were registered on 13 May 2021. Shortly after, on 17 August 2021, Germany requested an anti-arbitration injunction from the domestic German courts.[4] After inconsistent decisions at the lower level, the matter reached the BGH. The case is of interest by the way in which it approaches the EU law question and the relationship between EU law, Member States’ obligations and the ICSID Convention.

 

  1. The reasoning of the BGH

1.1.  Applicability of §1032(2) ZPO: admissibility of the anti-arbitration injunction under procedural German law

One could say that the case before the BGH would never have come about if Germany, like most Model Law jurisdictions, had not endowed parties prima facie bound to arbitration agreements with the possibility of requesting anti-arbitration injunctions from domestic courts through Article 1032(2) Zivilprozessordnung (ZPO).[5]

Pursuant to §1025(1) ZPO, provisions such as §1032 ZPO are applicable when the seat of the arbitration is in Germany.[6] Article 1025(2) ZPO extends the international competence of German courts also to situations in which the seat of the arbitration is “abroad” or “not yet determined”.[7] This hardly applies to ICSID arbitrations.[8]

This could have been enough for the BGH to dismiss Germany’s request. But the senior judges found that an “analogical application” of §1025(2) ZPO was in order,[9] thereby rendering it applicable (and, with it, enabling their own competence to issue anti-arbitration injunctions) “whenever there is no domestic seat”.[10]

1.2. EU law prevails over the competence-competence of ICSID arbitrators

The BGH agreed that the ICSID Convention constitutes a closed or self-contained legal system with its own procedural rules.[11]. It acknowledged the principle of competence-competence enshrined in Article 41 ICSID Convention.[12] From the moment the proceedings have been registered, ICSID arbitrators are themselves responsible for determining whether the dispute falls within or outside their jurisdiction.[13]

Simply, the BGH did not consider none of that to be a strong enough argument: the primacy of EU Law was incompatible with respecting the principle of competence-competence.[14] The BGH referred to the principle of autonomy of the EU legal order,[15] as well as to Romatsa, contending that if state courts are obliged to refuse enforcement of an ICSID award when the arbitration agreement is incompatible with EU law, then an anti-arbitration injunction ought to be issued as a way to ensure the full effectiveness of EU law “as soon as possible”.[16]

1.3. Inapplicability of Article 351 TFEU

Unlike the situation concerning the United Kingdom, it bears emphasizing that the BGH could not have avoided a conflict between the ICSID Convention and EU law by resorting to Article 351(1) (TFEU). Germany was a founding member of the European Economic Community in 1958, acceding to the ICSID Convention later in 1969.[17] Therefore, the BGH could not rely on pre-accession international obligations to determine the prevalence of the system created by the ICSID Convention.

  1. Shortcomings of the German approach

2.1. A case of anti-arbitration activism: the BGH provides unconvincing reasons as to the analogical application of §1025(2) ZPO

The BGH rightly acknowledges that ICSID arbitrations have a “delocalized” nature, thereby having no seat.[18] It accepts that ICSID arbitrations are decoupled from domestic legal orders, thus being a singular case of an “anational arbitration proceeding”.[19] The BGH’s own reading of the legal character and characteristics of ICSID arbitration (as delocalized and anational) thus constituted the strongest reason to find against an analogical application of §1025(2) ZPO. Simply, the court cannot – at the cost of contradicting itself – consider ICSID arbitrations as arbitrations with no seat and then decide to equate them for all purposes with “arbitrations with seat in Germany”, “arbitrations with seat abroad” and “arbitrations with not yet determined seat”.[20]

From this point of view, the BGH position inevitably presents itself as a case of anti-arbitration activism, in particular: anti-ICSID arbitration.

2.2. The BGH fails to convince as to why EU Law prevails over the principle of competence-competence enshrined in Article 41 ICSID Convention

The thinking of the BGH that the principle of effectiveness of EU law commanded it – as the court of a Member State – to ensure such effectiveness “as soon as possible” (meaning through an anti-arbitration injunction)[21] is understandable.

However, in my view, it is a position that can also be read as reinforcing the activist anti-arbitration stance of the BGH. The CJEU has not yet expressly offered caselaw to the effect that – where anti-arbitration injunctions are available as per the domestic procedural rules (as in Germany) – domestic courts are obliged to issue them even when that would represent a conflict with the competence-competence of arbitrators.

In the context of the case at hand, there is no clear indication that the CJEU’s caselaw has sought to sidestep the exercise of competence-competence by ICSID arbitrators. Contrary to the BGH’s position,[22] the CJEU’s dictum in the PL Holding ruling – pursuant to which Member States are obliged to object to the jurisdiction of arbitral tribunals whenever the arbitration agreement is incompatible with EU law and then eventually seek the annulment or non-enforcement of the award – does not point otherwise.[23]

The BGH’s reasoning also seems to conceal a logical leap.

Implicitly admitting that there is no express obligation to issue the anti-arbitration injunction, the BGH premises its reasoning on the CJEU’s caselaw concerning the enforcement of intra-EU ICSID awards, pointing especially to the Romatsa ruling.[24] The logical leap is to draw on the potential refusal of an enforcement to enable an anti-arbitration injunction and thus render the arbitrators’ competence-competence inoperable.[25]

The arguments invoked to sacrifice the principle of competence-competence are ALSO not convincing: (i) it is doubtful that the principle of competence-competence impairs the effectiveness of EU law,[26] since an incompatibility between an ICSID award and the EU legal order can be sorted out at the enforcement stage; (ii) the need for an early declaration of the inadmissibility of arbitration proceedings to thereby provide “EU law as soon as possible with full effectiveness[27] is also hardly necessary, since the arbitral tribunal might decline jurisdiction over the dispute in accordance with the CJEU’s caselaw; (iii) finally, it is irrelevant whether the anti-arbitration injunction declared by the BGH has the effect of preventing the “later enforcement of the ICSID award in Germany”, since – as the BGH recognizes – the CJEU’s Romatsa ruling appears to have settled that.

This means that the BGH may have voluntarily adopted a precautionary defence of EU law that was not legally warranted. Instead, to minimize the conflict between EU law and international law and to ensure an interpretation of domestic law in accordance with Germany’s international obligations under the ICSID Convention, the court could have left the matter to be tackled at the enforcement stage if and when an ICSID award had been issued in favour of the investor.

  1. Conclusion
 

The German approach to the enforcement of intra-EU ICSID awards is relevant and interesting for two main reasons.

First, at the domestic level, it has shown that the BGH did not shy away from finding enough procedural arguments to reach the point of being able to issue the anti-arbitration injunction requested by Germany. As noted, this was far from evident or necessary.

Second, it is nonetheless an important decision from an EU law perspective. One may disagree with the CJEU’s stance on intra-EU investment arbitration (and criticize its caselaw from many different points of view), but it cannot be denied that the CJEU has been consistent and effective in its purpose of dismantling it. The courts of Member States, bound as they are to EU law, do not have a lot of flexibility concerning the (non-)enforcement of intra-EU ICSID awards. The BGH may have been wrong in jumping to the anti-arbitration injunction, but it was not generally wrong on its observations about what EU law would ultimately command it to do at the ulterior enforcement stage (Romatsa made it clear). This has all perhaps one caveat: Germany was a founding member of the EU, so there was no room to give effect to the ICSID Convention through Article 351 TFEU (as the British courts did). Accordingly, the courts of Member States which have joined the EU after acceding to the ICSID Convention will be faced with the following question when deciding on a request for the enforcement of an intra-EU ICSID award: do the ICSID Convention pre-accession obligations or does the CJEU’s Romatsa ruling prevail?

[1] The views expressed in this blog post are those of the author solely and do not represent the views of the firm, nor do they bind the firm in any way.

[2] See, before the decision was issued, the short analysis of Alexander M. Wagner, “Will Achmea Trump ICSID in Germany? Bundesgerichtshof Likely to Grant “Anti-Arbitration” Declarations”, Kluwer Arbitration Blog, 10 June 2023 (available at: https://arbitrationblog.kluwerarbitration.com/2023/06/10/will-achmea-trump-icsid-in-germany-bundesgerichtshof-likely-to-grant-anti-arbitration-declarations/).

[3] ICSID Case No. ARB/21/26.

[4] BGH (I ZB/43/22), 27 July 2023, p. 6.

[5] Article 1032(2) ZPO: “(2) Bei Gericht kann bis zur Bildung des Schiedsgerichts Antrag auf Feststellung der Zulässigkeit oder Unzulässigkeit eines schiedsrichterlichen Verfahrens gestellt werden.”

[6] Article 1025(1) ZPO: “Die Vorschriften dieses Buches sind anzuwenden, wenn der Ort des schiedsrichterlichen Verfahrens im Sinne des § 1043 Abs. 1 in Deutschland liegt.”

[7] Article 1025(2): “Die Bestimmungen der §§ 1032, 1033 und 1050 sind auch dann anzuwenden, wenn der Ort des schiedsrichterlichen Verfahrens im Ausland liegt oder noch nicht bestimmt ist.”

[8] BGH (I ZB/43/22), 27 July 2023, p. 11.

[9] BGH (I ZB/43/22), 27 July 2023, p. 13.

[10] BGH (I ZB/43/22), 27 July 2023, p. 13.

[11] BGH (I ZB/43/22), 27 July 2023, p. 21 (“Das Kammergericht hat zutreffend darauf abgestellt, dass das ICSID-Übereinkommen ein geschlossenes Rechtssystem mit eigenen Verfahrensregelungen aufweist.”).

[12] BGH (I ZB/43/22), 27 July 2023, p. 19.

[13] BGH (I ZB/43/22), 27 July 2023, p. 22-23.

[14] BGH (I ZB/43/22), 27 July 2023, p. 25.

[15] BGH (I ZB/43/22), 27 July 2023, p. 25.

[16] BGH (I ZB/43/22), 27 July 2023, p. 27.

[17] BGH (I ZB/43/22), 27 July 2023, p. 32.

[18] BGH (I ZB/43/22), 27 July 2023, p. 12.

[19] BGH (I ZB/43/22), 27 July 2023, p. 13.

[20] BGH (I ZB/43/22), 27 July 2023, p. 14.

[21] BGH (I ZB/43/22), 27 July 2023, p. 27.

[22] See BGH (I ZB/43/22), 27 July 2023, p. 30.

[23]   CJEU, Case C-109/20 – PL Holdings, dated 26 October 2021, §52-53 (“Lastly, it follows both from the judgment of 6 March 2018, Achmea (C‑284/16, EU:C:2018:158), and from the principles of the primacy of EU law and of sincere cooperation, not only that the Member States cannot undertake to remove from the judicial system of the European Union disputes which may concern the application and interpretation of EU law, but also that, where such a dispute is brought before an arbitration body on the basis of an undertaking which is contrary to EU law, they are required to challenge, before that arbitration body or before the court with jurisdiction, the validity of the arbitration clause or the ad hoc arbitration agreement on the basis of which the dispute was brought before that arbitration body.”).

[24] See BGH (I ZB/43/22), 27 July 2023, p. 28.

[25] See BGH (I ZB/43/22), 27 July 2023, p. 28.

[26] See BGH (I ZB/43/22), 27 July 2023, p. 29.

[27] BGH (I ZB/43/22), 27 July 2023, p. 30.