Judicial dialogue in the EFTA pillar of the EEA – developments and challenges

This is the first of two blog posts analysing the relationship between the EFTA Court and the highest courts of the EEA/EFTA States. In the first blog, Prof Halvard Haukeland Fredriksen explains how judicial cooperation in the EFTA-pillar of the EEA differs from that found in the EU, and how this played out in the recent Fosen-Linjen case. The second post (to be published Monday 4 November) will offer some thoughts on the credibility of the institutional framework of the EEA and present some reforms proposals.

Author: Halvard Haukeland Fredriksen

Download blog as PDF

1. A partner-like relationship

The relationship between the EFTA Court and the highest courts of the EEA/EFTA States is complicated. For reasons of sovereignty, the EEA/EFTA States were unwilling to copy the hierarchical relationship that exists between the EU Court of Justice (CJEU) and the national courts of the EU Member States. In a significant manifestation of the pragmatism that underpins the entire EEA Agreement, the EU accepted this. As a result, the Main Part of the EEA Agreement acknowledges the role of the courts of last instance of the EEA/EFTA States in a way that finds no parallel in the EU treaties (Article 106 EEA). Furthermore, the EEA Agreement does not oblige the EEA/EFTA States to establish a procedure for preliminary references to the EFTA Court equalling the one found in EU law (now Article 267 TFEU). The EEA Agreement itself only mentions the possibility for an EEA/EFTA State to allow a court or tribunal to ask the CJEU to decide on the interpretation of an EEA rule (Article 107 EEA). Rather than opening up for preliminary rulings from the ‘foreign judges’ of the CJEU, however, the EEA/EFTA States strengthened the two-pillar structure of the EEA by establishing their own, ‘softer’, version of the preliminary reference procedure: Courts and tribunals of the EEA/EFTA States ‘may’ request the EFTA Court to give ‘advisory opinions’ on the interpretation of the EEA Agreement (Article 34 SCA). As noted by the EFTA Court in a case from 2012, the result is a more ‘partner-like’ relationship between it and the highest courts of the EEA/EFTA States (Case E-18/11).

At best, a partner-like relationship between courts leads to a genuine judicial dialogue, in which different perspectives and arguments contribute to more thorough (and thus persuasive) judicial decisions than the courts involved would be able to deliver on their own. This, however, presupposes that the courts are prepared not only to engage with each other through transparent reasoning, in which also relevant counter-arguments are shown to have been considered, but also to reconsider previous conclusions in the light of new legal material and new legal arguments brought forward by their dialogue partners. This is particularly so in cases where the courts of the EFTA-pillar of the EEA deal with parts of EEA law that are simply taken over from EU law, as the judicial dialogue in such cases is complicated by the CJEU’s role as the authoritative interpreter of EU law (Article 19 TEU). Unlike the courts of the EFTA-pillar, the CJEU can rely on its authority, something that may help explain why the reasoning in its preliminary rulings is not always as thorough and comprehensible as one would wish for. Gaps and inconsistencies in CJEU case law complicates the judicial dialogue in the EFTA-pillar of the EEA, but it also makes it more important. In a nutshell, the challenge is to come up with an interpretation of the common EU/EEA rules that will persuade also the CJEU if it is asked to rule on the matter in future.

2. Judicial dialogue in cases of potential judicial conflicts

For quite some time, the main obstacle to a genuine judicial dialogue between the EFTA Court and the highest courts of the EEA/EFTA States was the lack of referrals, in particular from the Supreme Courts of Iceland and Norway. With the laudable increase in such referrals in the last couple of years, however, attention has shifted to how the dialogue is, and ought to be, conducted, in particular in cases where the point of departure appears to be quite different views on the proper interpretation of EEA law.

At the centre of the current discussion stand two recent referrals from the Supreme Court of Norway, both of which essentially ask the EFTA Court for clarification and amplification, or possibly even reconsideration, of earlier advisory opinions. The first one, in the Fosen-Linjen case, caused the EFTA Court to clarify its view on the liability for public authorities for breaches of EU/EEA public procurement law in a way that to most observers came very close to a retreat (Case E-7/18). The second referral, in the Campbell case, questions the EFTA Court’s analogous application of the Citizenship Directive to situations outside the Directive’s reach as a matter of EU law. As the latter case is still pending before the EFTA Court (Case E-4/19), it will not be dealt with here, but for two remarks: The referral confirms that the Supreme Court takes its obligation under Norwegian law to assess for itself whether to follow an advisory opinion or not, seriously. It also suggests, however, that where a possible deviation is based on legal material not previously considered by the EFTA Court (in this case a number of new decisions from the CJEU), the EFTA Court will be given the opportunity to clarify its view. This is a clear improvement if compared with the (in)famous STX case from 2013, where the Supreme Court deviated from an advisory opinion obtained by the Court of Appeal, without giving the EFTA Court the opportunity to express its view on the at least partially new arguments and approaches that the Supreme Court found to be convincing.

3. The Fosen-Linjen saga

Turning to the Fosen-Linjen case, the new referral from the Supreme Court allowed the EFTA Court to defuse an otherwise imminent judicial conflict. The EFTA Court’s first advisory opinion in the case (Fosen-Linjen I) was interpreted by most (but not all) commentators as effectively overruling the Supreme Court’s long held view that public authorities may only be held liable for an aggrieved tenderer’s loss of profit (‘positive contract interest’) in cases of material breaches of the EU/EEA public procurement rules. In the opinion, the EFTA Court briefly mentioned the Supreme Court’s case law, but did not engage with it in its own reasoning.

Two circumstances allowed the Supreme Court, ‘in the interests of dialogue between the EFTA Court and the national courts’, to ask for ‘clarification and amplification, or possibly even reconsideration’ of the first opinion.

Firstly, the operative part of the first advisory opinion did not state that the strict liability read into the Remedies Directive applied also to claims for loss of profits. A contextual reading of the opinion suggested an affirmative answer, but it was possible to argue, as indeed the EU Commission later did in the second round, that the question of which heads of damage were to be subsumed under the strict liability was left for the EEA States to decide.

Secondly, some of the legal material brought before the Supreme Court appeared not to have been considered by the EFTA Court. The list included an Explanatory Memorandum from the Commission that revealed that it, anno 1990, certainly did not intend to harmonise contracting authorities’ liability for loss of profits (COM(90)297); two recent judgments from the Supreme Courts of the UK and Sweden, respectively, that supported the established case law of the Norwegian Supreme Court; and a recent judgment from the CJEU that confirmed the nature of the Remedies Directive as a minimum harmonisation instrument (Case C-300/17).

Unfortunately, the analysis of the Supreme Court’s decision to send the case back to the EFTA Court is complicated by coincidental changes of the EFTA Court’s composition. At the time of Fosen-Linjen I, the Icelandic judge of the EFTA Court was ill and thus replaced by an ad hoc judge. Then, just weeks after the opinion in Fosen-Linjen I, the court’s then president (and judge-rapporteur in the case) abruptly decided to step down. For a court with only three members, two new judges is a lot. In addition, the only member of the Fosen-Linjen I court that was expected to take part in a potential Fosen-Linjen II, the Norwegian judge, was reported by the press to have told the Chief Justice of Norway that the EFTA Court would not be ‘offended’ if the Supreme Court decided to make a new referral. Interpretations of this comment varied greatly, from a mere statement of the obvious (the EFTA Court has no business getting ‘offended’ just because a national court find it difficult to understand an advisory opinion) to an invitation of a rematch and even an indirect disclosure of disagreement with Fosen-Linjen I.

Given these events, it is to be stressed that no trace of them is to be found in the Supreme Court’s decision to refer the case back to the EFTA Court and, furthermore, that the grounds given in that decision themselves fully substantiate the action taken. In light of the new legal material presented by the Supreme Court, and its stated interest in a judicial dialogue with the EFTA Court, it is actually somewhat unfortunate that it was not the ‘Fosen-Linjen I court’ that was given the opportunity to clarify the first advisory opinion. As this was not an alternative, however, perhaps it was a blessing in disguise that Fosen-Linjen II ended up being decided by three new judges (as the Norwegian judge unfortunately fell ill shortly before the oral hearing and therefore had to be replaced by an ad hoc judge).

Before the EFTA Court anew, two important contributions facilitated the reconsideration of the first opinion. Firstly, the EFTA Surveillance Authority stood its ground from the first round and maintained its view that the Remedies Directive does not entail strict liability for loss of profits caused by a breach of the EEA rules governing public procurement. Secondly, in a remarkable development from its observations in the first round, the EU Commission offered a thorough textual, contextual, historical and purpose-oriented interpretation of the Directive that caused it to conclude that it was ‘inconceivable’ that the EU-legislator had intended to harmonise national contracting authorities’ liability for loss of profits. Unlike the EFTA Surveillance Authority, the Commission stopped short of stating that Fosen-Linjen I should be reconsidered, but its view that the Remedies Directive leaves it to the Member States to decide which heads of damage that fall under the strict liability advocated in that case, came very close to the same result.

Given the different interpretations of Fosen-Linjen I that had revealed themselves in the proceedings in Fosen-Linjen II, the EFTA Court had little difficulty rejecting the view that the second reference should be held to be inadmissible or, in the alternative, that the EFTA Court should simply reaffirm the first opinion by a reasoned order referring to Fosen-Linjen I. Still, in the context of judicial cooperation in the EFTA-pillar, it is noteworthy that the EFTA Court tacitly welcomed the new referral by reiterating that it is ‘important that questions on the interpretation of the EEA Agreement are referred to the Court under the procedure provided for in Article 34 SCA if the legal situation lacks clarity’, and reaffirming that all it takes for a second request to be admissible is for the national court to submit ‘new considerations which might lead to a different answer to a question submitted earlier’.

The EFTA Court then proceeded to clarify that the Remedies Directive is indeed an instrument of minimum harmonisation and, based on that finding, that it does not entail strict liability for loss of profits caused by a breach of the EEA rules governing public procurement.

The EFTA Court’s answer in Fosen-Linjen II made the job for the Norwegian Supreme Court a lot easier than it would if Fosen-Linjen I had been reaffirmed. Strikingly, before the Supreme Court neither Fosen-Linjen nor its supporter the Confederation of Norwegian Enterprise even attempted to rely on Fosen-Linjen I, arguing instead that the ‘sufficiently serious breach’-threshold should not be considered particularly high. The Supreme Court itself simply noted that Fosen-Linjen I ‘appeared’ to advocate for a strict liability, but that this had been ‘nuanced’ by Fosen-Linjen II, and then held that the liability of public authorities can be limited to cases of a sufficiently serious breach of the EEA rules governing public procurement. The Supreme Court proceeded to set out in greater detail the liability threshold, through reasoning distinguished by multiple references to both the EFTA Court and the CJEU. In replacing its own previous ‘material error’-test by the common EU/EEA ‘sufficiently serious’-test, the Supreme Court also confirmed its willingness to reconsider its case law in light of developments of EU/EEA law. As the government’s view of that test was rejected, and the breach in the case at hand found to be sufficiently serious, the judgment also confirms the Supreme Court’s standing as far from ‘State-friendly’ in cases concerning public authorities’ liability for breaches of EEA law.