Marc J. Goldstein Arbitrator & Mediator NYC
January 15, 2019

Suspended in Sweden: The Achmea Controversy Visits Washington

Let’s have a show of hands. All in favor of the following proposition:

             A U.S. District Court In An Award Enforcement Case Should Decide      If European Union Law Invalidates the Jurisdiction of An Arbitral Tribunal Under the Energy Charter Treaty

Too puzzled to cast a vote?  Then read on.

This question came to my attention when I heard at a recent arbitration conference that at least three, perhaps four, US district court judges (three in Washington, one in New York) have been asked recently to decide this question, and are either puzzling over it now or waiting for completion of the record.  A sensible colleague in the conference audience asked whether US district judges were up to the task, given its foreign law complexities. The panelist-presenter, evidently steeped in those complexities (arising from the interplay of EU law, the Energy Charter Treaty’s dispute settlement provisions, and a certain famous (or infamous) European Court of Justice (ECJ) judgment from 2018 called Achmea), worried aloud that they may not be. The question addressed in this post is whether the US district court in one of those cases – where the record seems to be rather complete and the motion to enforce the award evidently is rather fully submitted – should take up the suggestion of the award-creditor Kingdom of Spain to stay the proceedings, as the New York Convention permits the Court in its discretion to do where the Award has been suspended at least temporarily  by a competent court at the seat of the arbitration (Stockholm) and where an application to set aside the award has been made in that court.

Having spent a fraction of Holiday time binge-reading portions of the record in Novenergia v. Kingdom of Spain (U.S. District Court for the District of Columbia, Docket No. 18 Civ. 1148), I first offer a bit of background. The Energy Charter Treaty (ECT), signed or acceded to by 54 countries and the EU (emphasis supplied without prodding by Novenergia’s counsel), is a source of protection for foreign direct investment in the energy sector and provides for dispute settlement between Contracting States and investors of other Contracting States under one of several options for arbitration: ICSID Rules, ICSID Additional Facility Rules, UNCITRAL Rules ad hoc arbitration, or arbitration under the Rules of the Stockholm Chamber of Commerce. The US and Canada are ECT Observer States, not Contracting States, as are most of the major oil-producing nations in the Middle East.  Starting around 2004, the Kingdom of Spain, an ECT Contracting State, pursued a national commitment to renewable energy via solar power by providing investment incentives in the form of subsidies through the tariffs that solar power electricity producers could charge for feeding voltage into the Spanish grid. But when Spain began to realize that the subsidies were bankrupting the Kingdom, beginning around the start of this decade, the incentive program was significantly revised and constricted, causing grief for investors who had relied on the incentives, and, in short order, a number of ECT arbitrations against Spain by investors from other ECT Contracting States. (Novenergia is Luxembourgeois). Many of those cases were pending (and some still are) when the ECJ, the EU’s highest court, in March 2018 issued its Judgment in the Achmea case (which did not involve the ECT but rather a bilateral investment treaty (BIT) between two EU Member States). For reasons now explained, Achmea (coupled with European Commission guidance on its impact) gave Spain and other EU Contracting States who are Respondents in ECT arbitrations, an idea for a potential jurisdiction defense.

Now a few words about Achmea (Slovak Republic v. Achmea BV, Case No. C-284/16, ECJ Judgment of March 6, 2018, found, e.g., at the ECJ’s website Consider Achmea, US readers, to be the Marbury v. Madison of EU investment arbitration. It is emphatically the province of the European Court of Justice, the ECJ, to say what EU law is. Or at least that’s the view of the EU. They call this “The Autonomy of EU Law.” Generally, courts of EU Member States must apply EU law, which is intrinsic to the domestic law in each State. (Note to EU-aspirant States: when the EU sells admissions online, this intrinsicality takes hold when you click “Accept”). Arbitral tribunals under intra-EU BIT’s are creatures of those treaties, not organs of the judicial systems of the State parties to the BITs. The awards of such BIT tribunals are regulated by judicial review in courts at the seats of the arbitrations, which apply the arbitration law and the public policy of the seat State. In a given case under a given BIT this may or may not be a court of the Respondent Contracting State or even another EU Member State. Within the courts of EU Member States, but not elsewhere, and not before BIT arbitral tribunals, there is a procedure to obtain from the ECJ a preliminary ruling on a question of EU law that has arisen in the case. Achmea, a Dutch company and Slovakia chose to arbitrate in Frankfurt, and the German Bundesgerichtof  (that’s a Court) referred to the ECJ the question of whether intra-EU BIT arbitration was in violation of the Treaty on the Functioning of the European Union, given that the arbitral tribunal had neither the right nor the obligation to make such a referral to the ECJ. An arbitral tribunal under an intra-EU BIT doesn’t derive an obligation to apply EU law from the Respondent State’s EU subscription, but only from the BIT, which typically tells arbitrators to apply (inter alia) the BIT, general principles of international law, and law of the State party insofar as necessary. But that sets up intra-EU BIT arbitration as a forum where potentially an “off-shore” version of EU law could evolve, at odds with the official version emanating from the ECJ courthouse in Luxembourg. And in Achmea an aggrieved Slovak Republic  — which got its EU Membership Card roughly a decade after signing the relevant BIT with The Netherlands — argued that this is not good, that The Autonomy of EU Law negates the BIT’s standing offer to arbitrate with an investor of the counterpart Member State because the potential arbitral tribunal envisioned by the BIT is an illegitimate “off-shore” applier of EU law. No lawful offer to arbitrate = no arbitral jurisdiction = proper grounds to annul the award. The ECJ in Achmea agreed on the legal invalidity of the standing offer to arbitrate in an intra-EU BIT, and the German court that had referred the question to the ECJ then proceeded to annul Achmea’s award.

Depending on whether you are a victorious EU investor or a losing EU State in an ECT arbitration, Achmea means either everything, or nothing, in regard to the enforceability of the award. That is to say, if an ECT arbitral tribunal is an illegitimate applier of EU law in the same sense that an intra-EU BIT tribunal (according to the ECJ) is, then ECT arbitration between EU investors and EU Members cannot subsist.  Fascinating. But our question is whether this is a perfectly appropriate issue to be decided in the first instance by a US district court in Washington DC – ahead of the Svea Court of Appeal in Stockholm, already asked by Spain to annul the award, ahead of the ECJ, which could be asked by the Stockholm court to decide in a preliminary ruling whether Achmea extends to a multi-lateral treating that includes non-EU Members like the ECT, and despite a preliminary order of the Stockholm court (issued a mere two days after the filing of Spain’s annulment application) that suspends the award and thus bars its enforcement while all this gets sorted out.

US law guidelines for a district court’s exercise of discretion to stay or not stay the enforcement of a New York Convention award pending an annulment proceeding at the seat of the arbitration comes mainly from a US Second Circuit Court of Appeals case, Europcar Italia, S.p.A. v. Maiellano Tours, Inc., 156 F.3d 310 (2d Cir. 1998), where the Court identified a number of considerations that district courts should take into account:

  • The general objectives of arbitration – the expeditious resolution of disputes and the avoidance of protracted and expensive litigation;
  • The status of the foreign proceedings and the estimated time for those proceedings to be resolved;
  • Whether the award sought to be enforced will receive greater scrutiny in the foreign proceedings under a less deferential standard of review;
  • The characteristics of the foreign proceedings including (i) whether they were brought to enforce the award (which would tend to weigh in favor of a stay) or to set the award aside (which would tend to weigh in favor of enforcement); (ii) whether they were initiated before the underlying enforcement proceeding so as to raise concerns of international comity; (iii) whether they were initiated by the party now seeking to enforce the award in federal court; and (iv) whether they were initiated under circumstances indicating an intent to hinder or delay resolution of the dispute;
  • A balance of the possible hardships to each of the parties, keeping in mind that if enforcement is postponed under Article VI of the Convention, the party seeking enforcement may receive “suitable security” and that, under Article V of the Convention, an award should not be enforced if it is set aside or suspended in the originating country; and
  • Any other circumstances that could tend to shift the balance in favor of or against adjournment.

The Second Circuit went on to say that this was not “an exhaustive list,” and that “[b]ecause the primary goal of the Convention is to facilitate recognition and enforcement of arbitral awards, the first and second factors on the list should weigh more heavily in the district court’s determination.” Obviously this list of considerations was stated, and weighted, in the context of enforcement of a private commercial arbitration award, an award that resolved breach of contract issues that were unique to the parties’ commercial contract, had no intergovernmental or public policy significance, and carried no risk of deciding issues of public law that were also at the same time presented in a number of other investment arbitrations and in award enforcement or award annulment proceedings before foreign courts. Novenergia v.Spain thus presents a test of the adaptability of the Maiellano Tours formula to a radically different context.

The first Maiellano Tours factor – in a word, expedition – does not appear to favor a merits decision here and now. Rather we should be concerned here that haste may make waste – that if the ECJ formally embraces the EU position that the Achmea outcome extends to intra-EU disputes under the ECT, any judgment given by the US district court may have to be vacated, and any execution obtained against US-based assets of the Spanish State might need to be reversed. (I have no idea what prejudice in terms of effective execution might befall Novenergia from such a delay, but the US district court may hear the parties on that question and require suitable security). One is reminded of what occurred in the Southern District of New York and US Second Circuit Court of Appeals in the Thai-Lao Lignite case, where a stay of enforcement was denied despite anticipated annulment proceedings at the seat in Malaysia, the award was enforced, the award was then annulled in Malaysia, and several years of proceedings ensued here in New York concerning the vacatur, or not, of the initial enforcement judgment. And that happened in a situation where there was initially considerable doubt about the merits of the annulment application at the seat. Here, the parties’ respective Swedish law experts agree that the Award has been suspended under the Swedish Arbitration Act (albeit as a temporary measure and without elaboration of reasons), and the Swedish law expert for Spain, the retired Supreme Court of Sweden justice, states that such a suspension typically occurs only when the issuing court has determined that the annulment application has substantial merit. So, for the US district court in Washington DC to forge ahead now to the merits of controversial questions of EU law and ECT interpretation under international law runs some significant risk of extending rather than streamlining the judicial proceedings.

The second Maiellano Tours factor, status of the foreign proceedings, is really a refinement of the first “expedition” factor, asking how much delay is expected if completion, or at least material progress, is awaited in the foreign proceedings.  Here Novenergia’s expert states that the Stockholm annulment case could perhaps be finished by Summer 2019, but that it might take longer if counsel ask for more time or if the Stockholm court decides to seek a preliminary ruling from the ECJ on the Achmea/ECT issue. That expert also posits that the Achmea issue is of such importance that one can predict rather confidently that the Stockholm court (Svea Court of Appeal) will exercise its discretion to certify the case for appeal to the Supreme Court of Sweden, and that if this were to occur the end date for the entire Swedish annulment process could extend to 2022. Spain’s expert, a recently-retired Justice of the Supreme Court of Sweden, does not disagree.  Your Commentator’s view is that the time involved for the Swedish annulment process does not support denial of the stay. One element of this assessment is that the minimum time for an initial decision, more or less six months (and growing nearer while the matter is sub judice), is not so long. Another element is that the stay may be reconsidered, perhaps after the first Swedish decision, or even earlier, if and when there is a preliminary ruling from the ECJ on whether the Achmea ruling extends to intra-EU arbitrations under the ECT. If there is a clarifying ECJ ruling, adherence to that ruling by the Swedish courts may be foreseeable, so the time for the further proceedings in Sweden to be completed may then be not particularly relevant to the analysis. A relatively short (up to six months) wait-and-see period when there is great volatility on this issue in a number of European courts and in ECT arbitral tribunals has considerable merit.

The third Maiellano Tours factor is “whether the award sought to be enforced will receive greater scrutiny [in the foreign court] under a less deferential standard of review.” Here the first question may be what is the scope of review in the US Court of the arbitral tribunal’s rejection of Spain’s Achmea-based objection to jurisdiction? Without digressing into a lengthy discussion of this controversial corner of US arbitration law, let’s assume for argument’s sake that the US Court would not view the parties’ agreement to Stockholm Rules arbitration in Stockholm in the framework of Swedish arbitration law as a clear and unmistakable delegation of all power to decide arbitral jurisdiction issues to the arbitrators.  Let’s further assume that in Sweden an arbitrator’s arbitrability decision is judicially reviewed more or less de novo when this is  requested (IBA Arbitration Guide for Sweden, updated Jan. 2018, at p. 8, prepared by Robin Oldestam of Mannheimer Swartling). So differential deference to the tribunal’s Achmea arbitrability ruling may not be present. But what about the status of EU law as an intrinsic element of Swedish substantive law? If the ECJ declares that the Achmea rationale extends to the ECT, isn’t Sweden bound to apply the ECJ ruling whereas the US Court is not? This would seem to be effectively a wider scope of review in Sweden, a factor that favors a stay under the Maiellano Tours factors, on the basis that the risk of inconsistent outcomes is reduced by waiting.

I will not dwell here on the fifth Maiellano Tours factor “balance of hardships”, save to note: (1) that alleged economic hardship of a Claimant like Novenergia may be partially relieved by a requirement of suitable security, and would not be relieved by permitting enforcement if Spain could prevail upon a court to require Claimants to create reserves sufficient to satisfy its claims for recoupment if the Achmea issue were to be ultimately resolved in its favor; and (2) it seems an unnecessary hardship for Spain to be required to litigate the Achmea issue before multiple enforcing courts unconnected to the EU when an EU Member State court such as the Svea Court of Appeal might grant Spain’s request that a preliminary ruling be sought from the EU concerning the application (or not) of the Achmea rationale to the Energy Charter Treaty.

The sixth (and last enumerated) Maiellano Tours factor is the “catch-all”: “[A]ny other circumstances that could tend to shift the balance in favor of or against adjournment.” One such circumstance is that Novenergia v Kingdom of Spain is only one of many intra-EU arbitrations under the Energy Charter Treaty. Conferees at the December 2018 conference that I attended were told that there are at least four award enforcement cases in US district courts at this time in which an investor from one EU Member State has won an award against another EU Member State for violations of the ECT. How many other similar ECT arbitrations are in progress I do not know, but indications are that there may be a considerable number. A Norton Rose Fulbright report in October 2018 states that Spain had been named as the Respondent in more than 40 cases, that Italy faces ten cases, that the most frequently occurring nationality of ECT Claimants is German, and that 66 ECT cases are pending. It stands to reason that many of these cases involve the Achmea issue, and that definitive guidance from the ECJ will be obtained, if not in the Novenergia case in the Swedish annulment case, then in another annulment case where an ECJ Preliminary Ruling is similarly sought. I suggest it is reasonable for a US district judge to tread cautiously because the stakes are far greater than the award in favor of Novenergia or the timing of its collection. Essentially the further utility of the ECT arbitral dispute resolution process within the EU is at stake. A merits decision on the issue by a US court in one case is not likely to resolve the matter and there is at least some risk that the US court, presumably having no prior training or experience in the relevant EU law, would issue a decision which, even if soundly reasoned, fails to be influential among the Justices of the ECJ when a Preliminary Ruling is predictably sought by a court of an EU Member State.

And also in the “other circumstances” category: What should be the influence on the US district judge of the fact that arbitral tribunals composed of renowned experts in public international law, including the Tribunal in Novenergia, have sided with Claimants in rejecting the position that Achmea’s reasoning should apply to intra-EU arbitrations under the ECT?  It would seem that there may be strong reasons for the US judge to resist (for now) the temptation to follow the lead of the investment arbitration “in crowd.” What of the fact that because the US is not a Member State of the ECT, the ECT is not American law, and US courts have neither any developed ECT jurisprudence nor do they have any direct forum interest in construing its dispute settlement provisions? What of the fact that the US is also not a Member State of the Vienna Convention on the Law of Treaties (“VCLT,” although the US is a non-ratifying signatory) — while Sweden is — and that it is through the VCLT’s principles of interpretation that the proposed construction of the ECT in light of EU law is to be made? What of the fact that the Svea Court of Appeal or the Supreme Court of Sweden might view an EC position (and/or ECJ Ruling) against intra-EU arbitration under the ECT as fostering a mandatory rule of Swedish substantive law, having the status of fundamental public policy, and thus requiring annulment of an Intra-EU ECT award rendered in Sweden — even if that conclusion would not be reached when analyzing the legal validity of Spain’s offer to arbitrate under the ECT solely on the basis of Swedish arbitration law? Doesn’t the suspension of the Novenergia award by the Svea Court of Appeal sufficiently hint at this possibility that there would need to be very compelling reasons favoring immediate US enforcement, even at the risk of later vacatur of the enforcement judgment, in order for a US court to deny a stay?

In summary, it would appear that there are substantial reasons for the US court to refrain at least temporarily from deciding the merits of the Novenergia v. Kingdom of Spain enforcement case to await developments in the annulment proceeding in Sweden.

Leave a Reply