By Benedikt Pirker
Should EU secondary legislation be reviewed against the benchmark of the provisions of an international agreement? In 2012 the General Court answered this question in the affirmative and annulled two decisions of the Commission which were based on a regulation which was deemed incompatible with the Aarhus Convention. However, the EU institutions appealed against those judgments. Consequently, in cases C‑401 to 403/12, Council e.a. v. Vereniging Milieudefensie and C-404 and 405/12, Council v. Stichting Natuur en Milieu e.a., the Grand Chamber of the Court was confronted with the same question. There is already quite some case law on the topic of review of legality within the EU legal order in light of international obligations of the EU, typically with the Court being hesitant to undertake such review. In the cases involving the Vereniging Milieudefensie and the Stichting Natuur en Milieu, the General Court and the Advocate General made, in my view, some valuable suggestions in favour of reviewing EU law against international agreements. Unfortunately, the Court decided to stick to its guns, thus continuing in the line of its own previous jurisprudence, and annulled the General Court’s judgments. The result leaves a somewhat sour taste for those who think that EU institutions and their legal acts should be amenable to judicial review under reasonable conditions. Not only is the very purpose of the EU regulation at issue to implement the obligations arising from the Aarhus Convention, but the Grand Chamber’s view also leads to a lacuna in legal protection in EU law exactly where the central aim of the Aarhus Convention would in theory be to provide individuals with access to justice.
Before analysing the reasoning of the General Court, the Advocate General and the Court itself, some legal background needs to be set out. The Aarhus Convention is an important international environmental convention which contains mainly three ‘pillars’: (i) access to information; (ii) public participation in decision-making; and (iii) access to justice in environmental matters. The EU acceded to the Convention in 2005. The Aarhus Regulation 1367/2006 deals with the application of the provisions of the Aarhus Convention to the institutions and bodies of the EU. As a central issue, Article 10 (1) of the Regulation provides that non-governmental organisations meeting certain criteria are entitled to request an internal review to the EU institution or body that has adopted a certain administrative act under environmental law or, in case of an alleged administrative omission, that should have adopted such an act. Article 2 (1) defines an ‘administrative act’ as ‘any measure of individual scope under environmental law, taken by a [EU] institution or body, and having legally binding and external effects’, thereby excluding acts of a general scope from such internal review initiated by e.g. environmental NGOs. The provision can be understood as implementing the obligations resulting from Article 9 (3) of the Aarhus Convention. Article 9 (3) of the Aarhus Convention is part of the provisions on access to justice and reads that ‘[…] each Party shall ensure that, where they meet the criteria, if any, laid down in its national law, members of the public have access to administrative or judicial procedures to challenge acts and omissions by private persons and public authorities which contravene provisions of its national law relating to the environment’.
The following legal disputes between the EU institutions and a couple of Dutch environmental organisations mainly arise from the visible tension between the broad range of ‘acts and omissions’ which violate domestic environmental law for which the Aarhus Convention opens access to review and the rather limited notion of ‘measures of individual scope’ foreseen in the Aarhus Regulation.
In order to fully understand the actual proceedings and the dispute over the possible effects of the Aarhus Convention in the EU legal order, some context needs to be set out on the Court’s case law on the effect of international agreements in the EU legal order.
The Court’s case law on the effect of international agreements
Over the years, the Court has developed a complex jurisprudence on the topic of the effect to be given to the provisions of international treaties within the EU legal order. As one central problem, there is a somewhat fuzzy distinction between the terms of direct effect and review of legality that should be explained at least rudimentarily at this point. In a number of cases, the Court held that individuals could only invoke such provisions which gave them rights before courts if the agreement at issue could as such be seen as capable of having direct effect and if the provision at issue was sufficiently precise and unconditional. The example of WTO case law is perhaps best known: The Court insisted on the political flexibility which in its view WTO law grants to generally refuse direct effect. However, the Court also developed two exceptions where individuals could indeed rely on provisions of WTO law to challenge provisions of EU law. The Court thus held that it would review the legality of an EU act in the light of WTO law where the EU act at issue referred explicitly to specific provisions of WTO law (the Fediol exception) or where the EU intended to implement a particular obligation assumed under WTO law (the Nakajima exception). Already at some points in the Nakajima decision, but much more explicitly in Biotechnology the Court started to introduce a distinction between a situation where an international treaty created directly effective individual rights, i.e. direct effect in a narrow sense, and where it could nonetheless be used more broadly by courts to review the EU’s compliance with its obligations. In Biotechnology, the Court also held that the Rio Convention on Biological Diversity was, unlike WTO law, not based on ‘reciprocal and mutually advantageous arrangements’. This could be seen as meaning two positive things: first, the WTO law case was a special case and the Court’s unwillingness to grant the possibility of review of legality seemed to be limited to this regime of international law. Second, the distinction drawn between direct effect and the possibility to use provisions of an international treaty for the review of legality of EU acts even where strictly speaking these provisions had no direct effect also meant a higher potential for international law to be granted at least some effect in the EU legal order. However, in Intertanko the Court applied a similarly restrictive reasoning to a non-WTO international agreement, the United Nations Convention on the Law of the Sea (UNCLOS). It held that the Convention does not ‘in principle’ grant independent rights and freedoms to individuals and decided that therefore the ‘nature and broad logic’ of the Convention prevented the Court from assessing the legality of EU acts in the light of its provisions. First, the Court thereby denied any effect as a benchmark for review to an international treaty as a whole because of the latter’s structure. What is more, it also did not use its earlier distinction between direct effect and review of legality, but based itself exclusively on the fact that the provisions of UNCLOS did not confer rights to individuals. In Air Transport Association of America, the Court allowed review of EU legislation in light of an international treaty, but also merely because the ‘nature and broad logic’ of the agreement did not preclude this and because the agreement created directly and immediately applicable rules which conferred rights and freedoms to individuals, in this case airlines. There seems to be thus a shift away from the distinction between direct effect and review of legality and back to the earlier classic direct effect case law, i.e. a mere examination whether an agreement does not as a whole preclude any effect as a benchmark for review in the EU legal order and whether the provision to be applied regulates individuals in a sufficiently direct way. It is in this context that the question of the effect of the provisions of the Aarhus Convention in the EU legal order has to be understood.
In Lesoochranárske zoskupenie, the Court already had to discuss the issue of the effect of Article 9 (3) of the Aarhus Convention and held that the provision did not contain a clear and precise obligation capable of directly regulating the legal position of individuals. This already narrows down the road to direct effect in the sense of the respective case law of the Court. Centrally, Article 9 (3) refers to criteria to be laid down in national law, which makes it clear that a subsequent measure is required for the implementation of the provision. The two later disputes to be discussed now raise the question whether there is any other avenue which may lead to the review of legality of the Aarhus Regulation against the benchmark of Article 9 (3) of the Aarhus Convention.
The proceedings before the General Court
In the Vereniging Milieudefensie case, the Commission had previously accepted that the Netherlands postponed the deadline for attaining annual limit values for nitrogen dioxide in certain zones. The Vereniging Milieudefensie submitted a request for internal review of that Commission decision. The request, however, was subsequently rejected by the Commission because the decision was not a measure of individual scope and could not form the subject of an internal review procedure. In its judgment, the General Court decided that it could review the provision on internal review, Article 10 (1) of the Aarhus Regulation, in the light of Article 9 (3) of the Aarhus Convention. It based this on the fact that Article 10 (1) of the Regulation was an implementation of Article 9 (3) of the Convention and expressly referred to the latter (para. 58). It then found that the Aarhus Regulation did not correctly implement the obligations arising under the Aarhus Convention, because it unduly limited the possibility of review to acts of individual scope.
In the second case, Stichting Natuur en Milieu requested that the Commission carry out an internal review of Regulation 149/2008, a regulation that amends an earlier regulation setting certain maximum residue levels for specific products. The Commission declared the requests for internal review inadmissible under Article 10 (1) of the Aarhus Regulation because the regulation was not an ‘administrative act’ as required by Article 10 (1). The General Court, however, annulled this decision based on a similar reasoning on the incorrect implementation of Article 9 (3) of the Aarhus Convention by the Aarhus Regulation. Centrally, it found that the adoption of Regulation 149/2008 fell within the scope of the Commission’s regulatory activities and was not to be considered as ‘legislative’ activity, legislative activity being excluded from Article 9 (3) of the Aarhus Convention because of its general scope (para. 70).
The reasoning of the General Court thus follows the WTO case law of the Court and thereby the classic direct effect line of case law. However, rather than denying direct effect right away based on this case law, the General Court uses the previously mentioned Nakajima exception to find that the contested provision of the Aarhus Regulation implements Article 9 (3) of the Aarhus Convention. This paves the way for a review of legality of the regulation after all and the finding that the regulation is partially incompatible with the Convention. The General Court then annulled the decisions of the Commission based on the incompatible regulation. This contradicts, of course, the Court’s findings in Lesoochranárske zoskupenie, at least if the view is taken that the review of legality is inextricably linked to direct effect.
The Advocate General’s view
In the appeal cases initiated by the EU institutions, a different approach is suggested by the Advocate General, although it arguably would have produced a comparable outcome. In Vereniging Milieudefensie Advocate General Jääskinen distinguished the Aarhus Convention from the WTO treaties as being ‘by its nature […] a procedural instrument’ (para 87), a codification of procedural rights in relation to the environment, and not based on reciprocal and mutually advantageous arrangements like the latter (para 88). With regard to Article 9 (3) of the Convention, the Advocate General admitted based on Lesoochranárske zoskupenie that the provision requires the adoption of subsequent acts and that therefore individuals could not rely on it. However, he found it to be at the same time a ‘mixed provision’ since it also contained an obligation of a contracting party to ensure a ‘clearly identifiable outcome’ (para 92). The national legislature may thus possess discretion to determine criteria to be satisfied by organisations to be able to challenge an infringement of environmental law; however, the very obligation to guarantee access to justice was, for the Advocate General, sufficiently clear to preclude a rule with the effect of excluding certain categories of non-legislative decisions taken by public authorities from the possible scope of review (para 94). Therefore, based on ‘its objective and its broad logic’ Article 9 (3) of the Aarhus Convention could serve as a benchmark for the review of legality for the Advocate General. While both the General Court and the Advocate General follow thus generally the line of the direct effect case law, the General Court suggests relying on a pre-defined exception, while the Advocate General prefers to develop a doctrine of partial direct effect.
In Stichting Natuur en Milieu, by contrast, Advocate General Jääskinen found that the Commission acted in a legislative capacity when adopting the disputed Regulation 149/2008, mainly because there were no rules of environmental law which could restrict the Commission’s competence and serve as a benchmark against which the review procedure introduced by the Aarhus Convention could have been carried out (para 63). The adoption of Regulation 149/2008 did therefore not fall within the scope of the Aarhus Convention for the Advocate General.
The Court’s Grand Chamber decision
The Court found itself between the claim of the General Court to allow review of legality under the Nakajima exception and the Advocate General’s suggestion to find a directly applicable core of Article 9 (3) of the Aarhus Convention in Vereniging Milieudefensie (the decision in Stichting Natuur en Milieu is identical in the relevant section and therefore requires no separate discussion). However, the Court chose a third way out following its previous, direct effect based case law. While it did not explicitly base its decision on direct effect, it held that the provisions of an international agreement could only be relied upon to review an act of EU secondary legislation where ‘the nature and broad logic’ of that agreement did not preclude it and, secondly, the provisions at issue were, ‘as regards their content, […] unconditional and sufficiently precise’ (para 54). The Court then repeated its finding from Lesoochranárske zoskupenie that a subsequent measure was needed to implement Article 9 (3) of the Aarhus Convention.
In addressing the Fediol and Nakajima exceptions the General Court had brought up, the Court interpreted both of them restrictively as being ‘justified solely by the particularities of the agreements that led to their application’ (para 57). It thus found Fediol irrelevant, because Article 10 (1) of the Aarhus Regulation did not directly refer to specific provisions of the Aarhus Convention ‘nor confers a right on individuals’ (para 58). The Court then held that the Nakajima scenario was inapplicable because in that case the EU had created an implementing act in the ‘extremely dense’ antidumping system in accordance with existing international obligations, while the implementation of Article 9 (3) of the Aarhus Convention still left a broad margin of discretion when defining the rules for the implementation of the ‘administrative or judicial proceedings’ (para 59). In particular, the EU could not be considered to have intended to implement the obligations from Article 9 (3) of the Convention since the mentioned administrative or judicial procedures fell at the current state of EU law primarily within the scope of Member State law (para 60). Consequently, the Court found that the General Court had erred in reviewing the Aarhus Regulation against the benchmark of Article 9 (3) of the Convention.
The Grand Chamber’s point of view is in my view rather disappointing. One may give it the credit of being consistent with the Court’s earlier case law, but nonetheless it remains difficult to see where the road the Court is travelling down is supposed to lead. One of the ‘official’ aims of the EU is to contribute ‘to the strict observance and the development of international law’ (Article 3 (5) TEU). However, the Court has been unwilling to grant judicial review of EU law in light of international obligations that are – after all – beyond any doubt binding on the EU. It has done so since the days of the GATT 1947, repeated this position with WTO law and increasingly seems to find it acceptable also for other international treaties. The review of legality line of case law seems to have all but disappeared, to the benefit of the direct effect line of case law. Now the Court seems to be additionally tightening the conditions of the latter line so that review of EU law against the benchmark of international agreements in the EU legal order becomes the exception rather than the rule. One may start to wonder whether we are generally in a phase of withdrawal rather than engagement of the EU legal order with international law. Hardly ever does this become more obvious than in the present case, as two elements of the Court’s decisions show.
First, the Court rejects to apply the Nakajima exception. It seems to continue to narrow down the concept of EU law implementing international obligations to a degree where it is hard to see where this exception should ever apply. If the Aarhus Regulation does not implement the Aarhus Convention, what does it do then? Of course, Article 9 (3) of the Aarhus Convention leaves discretion and part of its implementation lies ‘primarily’ within the scope of Member State law. But there are obviously also implementing competences that the EU exercises, secondary legislation that is being put in place. So why should it be an all-or-nothing choice, i.e. why should there be no review at all for EU legislation if there is joint implementation of such international environmental obligations along the respective lines of competences of the EU and the Member States?
Second, the Court disregards the suggestion of the Advocate General to see Article 9 (3) Aarhus Convention as a ‘mixed provision’ and to allow review of EU secondary legislation against the benchmark of the clearly identifiable content of such a provision. Based on the Advocate General’s approach, it would have been possible to give some teeth to the requirement to provide access to justice. The Court’s approach has the awkward result that it denies even the possibility to review EU legislation against the benchmark of the provisions of the Aarhus Convention… on access to justice. The Court thereby allows the EU institutions to implement the Aarhus Convention and its central pillar of access to justice in a way that excludes a number of legal acts, in my view, unduly from any effective review. It remains unclear why exactly the fear of giving too much weight to international obligations should outweigh the important obligations at issue in the case.
As a side-effect, the Grand Chamber’s point of view also strengthens the impression that EU institutions can expect a somewhat more favourable treatment when EU secondary legislation is reviewed by the Court. As the Advocate General points out, too (para 132), the Court has generally vigorously protected the effectiveness and objectives of the Convention in its case law on the implementation of the Convention’s obligations by Member States; this approach is, however, only difficult to reconcile with the lenience now shown towards the case of EU implementing legislation.