Monday 23 January 2023

Anonymity in CJEU cases: the Court changes its approach


Peter Oliver, Honorary Professor at the Université Libre de Bruxelles


Photo credit: Isaac Castillejos   


Some readers may recall that in June 2018 the Court of Justice issued a press release, to the effect that, in keeping with the General Data Protection Regulation (GDPR), the Court had taken the following decision:


In order to ensure the protection of the data of natural persons involved in requests for a preliminary ruling while guaranteeing that citizens are informed and have the right to open courts, the Court of Justice has therefore decided, in all requests for preliminary rulings brought after 1 July 2018, to replace, in all its public documents, the name of natural persons involved in the case by initials. Similarly, any additional element likely to permit identification of the persons concerned will be removed.

                                                                                                (emphasis in the original)


This approach was to be followed not only in relation to the names of cases but would also apply to ‘all publications made as part of the handling of the case, from its lodging until its closure (notices to the Official Journal, Opinions, judgments...)’.  The names of individuals were to be replaced by two letters in each case, which were not their actual initials; and this would be supplemented by a few words in brackets briefly describing the subject-matter of the case. 


For the avoidance of doubt, the press release stated that the names of legal persons would continue to be used.  Another point which emerged clearly from the press release was that the new guidelines would apply only to preliminary rulings, the reason being that the General Court had no intention of falling into line with the higher court (but this reason was not spelt out in the press release). 


Furthermore, the Court expressly reserved the right to depart from this practice if “the particular circumstances of the case so justify” (a nebulous test if ever there was one) and “in the event of an express request from a party”. 


The new approach went well beyond the pre-existing practice of the Court of Justice and the General Court of replacing the names of natural persons in sensitive cases (e.g. where a litigant is a child and/ or an asylum seeker) by initials.  Manifestly, that is necessary to protect the privacy of the individuals concerned in keeping with Article 8 ECHR and Article 7 of the Charter.


In July 2018, the Court’s “Recommendations to national courts and tribunals in relation to the initiation of preliminary ruling proceedings” appeared in the Official Journal.  In two passages of these recommendations, national courts are encouraged to anonymise the names of natural person who are party to proceedings, when drafting the preliminary reference.  But of course the Court has no power to require the national courts to do so.


Those recommendations are supplemented by a notice on the Court’s website variously entitled “Anonymity in judicial proceedings before the Court of Justice” and “The protection of personal data in connection with publications relating to judicial proceedings before the Court of Justice”.  As to judicial proceedings before the Court other than preliminary references, the notice simply states that a person (who may not necessarily be one of the parties to the case in question) may apply to the Court to be granted anonymity.  With respect to preliminary references, the Court begins by recalling that, according to Article 95(1) of its Rules of Procedure, it is required to respect decisions of referring courts to grant anonymity to a party to the proceedings.  For the rest, the notice repeats the key passages of the Court’s press release of June 2018. 


What is more, the notice on the Court’s website points out that the policy change announced in the press release of June 2018 pre-empted the adoption of Regulation 2018/1725 of the European Parliament and the Council on the protection of national persons with regard to the processing of personal data by the Union’s institutions and bodies.  According to recital 5 in the preamble to this Regulation, it is intended to align the rules applicable in that context to the GDPR “as far as possible”.  The Regulation is expressed to apply to “all” the Union’s institutions and bodies (Article 2(1)).  Crucially however, the notice on the Court’s website does not suggest that the Regulation would preclude the Court from reverting to its practice prior to July 2018, should it wish to do so.  


In December 2018, I wrote a post on this subject which appeared on this blog.  In the post, I indicated what I believed (and still believe to be a number of shortcomings of the scheme set out in the press release of June that year.  In particular, I pointed out that this scheme would make it harder for courts, practitioners, legal academics and students to find, identify and remember the names of cases.


On 9 January 2023, the Court issued press release 1/23 announcing that as from 1 January the names of national persons in new preliminary references would no longer be composed of initials, but of fictional names which will not in principle be existing names.  These computer-generated fictional names are to be created by “dividing words into syllables, which are then randomly combined to produce fictional names”.  A generator has been created “for each official language of the European Union and additional generators will be developed, where necessary, for languages of third countries”. 


The press release also contains the following statement:


The allocation of fictional names does not affect:


-  References for preliminary rulings in which the name of the legal person is sufficiently distinctive (the name of that legal person will be used as the name of the case);

-  Direct actions (the Court of Justice will continue to allocate a conventional name to those cases, which will appear in brackets after the usual name of the case);

-  Requests for opinions;

-  Appeals;

-  Cases before the General Court.


In relation to the problem mentioned above, the Court’s press release is most welcome: the new scheme will undoubtedly make it easier for all concerned to find and remember case names.  However, it fails to solve the other issues mentioned in my earlier post.


First of all, each individual’s fundamental right under Article 8 and 10 ECHR and Articles 7 and 9 of the Charter to be known by his or her own name will still not be respected.  The Court’s stated willingness to accede to an “express request” from a party who is a natural person for the use of his or her own name is an empty letter unless that person is informed of the true position at an extremely early stage after the preliminary reference is made.  After all, it is entirely reasonable for the individual, who does not happen to be acquainted with the minutiae of the Court’s procedure, to assume that the Court will use precisely the same case name as the referring national court.  Moreover, the Court’s readiness to depart from its usual practice of imposing anonymity where “the particular circumstances of the case so justify” is of little or no avail to such an individual, since by definition recourse to this exception is a entirely a matter for the Court’s discretion.


Second, the Court’s decision not to respect the case names employed by national courts in preliminary references runs counter to the Court’s constant insistence on the fact that its co-operation with those courts in such proceedings is of the essence.  In any case, the Court has no means to coerce national courts to follow its practice, even if that were desirable.


For all these reasons, it is submitted that the Court should respect the choice of national courts except in highly exceptional cases.


Finally, by using computer-generated names the Court risks embarrassment or even ridicule.  Such names may well have unfortunate connotations of a racist, sexist or other inappropriate nature in the language of one or more Member States.  To avoid this, the Court’s administration will need to check that each computer-generated name is unobjectionable – at least in the language of the case and the language or languages of the parties.  This is likely to take time, which is unfortunate given the time pressure on the Court after it receives a preliminary reference.  However, this is obviously a minor concern.


In conclusion, whatever else may be said about the policy announced by the Court in 2018, the fact is that it still does not appear to have gained much traction: the General Court is still unwilling to follow suit, and it is not clear that many national courts have been persuaded to anonymise names in preliminary references.  Whether the Court’s change of tack now will garner much support remains to be seen.  In any event, the case for reverting to the Court’s practice prior to the summer of 2018 remains as strong as ever.


Wednesday 18 January 2023

Proposed AI Liability Directive: The EC lending a helping hand



Ida Varošanec (PhD student, University of Groningen) and Nynke Vellinga (post-doc researcher, University of Groningen)


Photo credit: Cryteria, via Wikimedia commons



1. Objectives of the proposal


On 28 September 2022, the European Commission published a proposal for an AI Liability Directive and an accompanying update of a complementary Product Liability Directive. In the preceding Report on Artificial Intelligence Liability, the Commission acknowledged the immense potential of artificial intelligence (AI). However, it has also identified the risks associated with it. For instance, connectivity of an AI-encompassing product can compromise its safety for users as it may be susceptible to cyber-attacks. Moreover, the outcomes of AI cannot always be predicted. To this end, ex ante risk assessments can be insufficient to address the possible wrongs. The opacity inherent in advanced AI-based products and systems makes it difficult to ascertain the responsibility of AI systems’ behaviours and choices. It is pivotal that humans can be able to understand how algorithmic decisions were reached in order to make a liability claim. Particularly, the opacity of AI systems can hinder victims in proving fault and causality in such cases. Consequently, the AI Liability Directive aims to ensure the provision of protection for victims of AI commensurate with those where damage has been caused by other products. It aims to increase trust in new technologies as well as to contribute to the ‘rollout of AI’ and improve its development in the internal market by preventing fragmentation and increasing legal certainty through harmonisation. Once adopted, these proposals will complement other AI regulation (e.g. the proposed AI Act) and establish liability rules for software and AI systems in the EU.


2. The proposed AI Liability Directive: scope


Other than what the name might suggest, the proposed AI Liability Directive does not provide any new ground of civil liability. That remains a matter of the national legislature, except when it comes to liability for defective products under the regime of the Product Liability Directive. Instead, the proposed AI Liability Directive provides for burden of proof rules on disclosure of evidence and rebuttable presumptions on a causal link. The rules cannot be invoked in every tort law case: only cases on fault liability fall within the scope of the AI Liability Directive. These are cases where liability for damage caused by (the use of) an AI system are based on fault. Fault encompasses wrongful actions and omissions. Due to the characteristics of AI systems, it can be difficult or prohibitively expensive to prove fault. Consequently, those suffering damage caused by an AI system might not get compensated for damage suffered, whereas those suffering damage from a non-AI system would be able to get compensated as they do not incur the same difficulties in proving fault. The proposed AI Liability Directive would address this discrepancy by providing rules on:


‘(a) the disclosure of evidence on high-risk artificial intelligence (AI) systems to enable a claimant to substantiate a non-contractual fault-based civil law claim for damages;

(b) the burden of proof in the case of non-contractual fault-based civil law claims brought before national courts for damages caused by an AI system.’ (art. 1 Proposal)


The proposed AI Liability Directive does not apply to risk-based liability claims. However, the proposed new Product Liability Directive does provide similar rules on disclosure of evidence and the burden of proof (art. 8 and 9).


The scope of the applicability of the proposed AI Liability Directive is partially limited to a specific category of AI system: the high-risk AI systems. For the definition of a high-risk AI system, the AI Liability Directive refers to the proposed AI Act. The AI Act identifies and lays down rules as per levels of risk associated with AI systems – those that carry (1) unacceptable risk, (2) high risk, and (3) limited risk. The fourth category – that of systems that pose a minimal risk (e.g. spam filters) – although within the material scope are not subject to any concrete rules. The first category – (1) unacceptable risk – concerns AI systems that are a clear threat to the safety, livelihoods and rights of persons (e.g. manipulation and social scoring systems). The third category (those of limited risk), is subject to specific transparency obligations due to their nature (e.g. deep fakes). High-risk AI systems represent those which are embedded in products subject to third-party assessment under sectoral legislation, and those which are not components of products but are deemed to be high-risk when used in certain areas (e.g. transport, education, safety components etc.). Such systems are subject to a set of requirements (e.g. risk assessments, mitigation systems, data quality, logging, and technical documentation) before being placed on the market.


The rules on disclosure of evidence as laid down in art. 3 of the proposed AI Liability Directive only apply to these high-risk AI systems. The rules on the burden of proof, however, apply to claims relating to all AI systems. (art. 4).


3. The proposed AI Liability Directive: disclosures and presumptions


3.1 Rebuttable presumption of a causal link


Article 4 introduces a rebuttable presumption of a causal link in the case of fault. It allows the courts to presume the causal connection between the fault of the defendant and the AI output (or failure to produce it) under three cumulative conditions. Firstly, the fault needs to be established (either by an assuming court or a claimant) consisting of non-compliance with the duty of care under EU or national law. Secondly, it must be ‘considerably likely’ that the fault has influenced the output of an AI system. Finally, damage by an AI system needs to be demonstrated. Paragraphs (2) and (3) differentiate between providers and users of AI systems.


The causal link concerning a claim for damages caused by a high-risk AI system shall not be presumed if the defendant demonstrates that sufficient evidence and expertise is reasonably accessible for the claimant to prove this causal link (art. 4(4)). When the claim concerns an AI system that is not high-risk, the presumption of the causal link shall only be applied where the national court considers it excessively difficult for the claimant to prove the causal link (art. 4(5)). Moreover, the defendant can always rebut any presumption regarding the causal link (art. 4(6)).


3.2 The disclosure of evidence


Article 3 of the proposed AI Liability Directive establishes the conditions regarding the disclosure of evidence and introduces a rebuttable presumption of non-compliance. This applies to high-risk AI systems as defined in the AI Act.


Article 3(1) of the Directive allows a court to order the disclosure of relevant evidence about specific high-risk AI systems that are suspected to have caused damage. Recital (16) confirms that this requirement has been unaccounted for by the AI Act proposal. However, the disclosure provided for in the AI Liability Directive does not seem to be absolute. Rather, it seems to be subject to a certain proportionality assessment since disclosure is only allowed to the extent necessary for sustaining the liability claim. To do that, national courts ought to consider the legitimate interests of all parties. Particularly, this applies in relation to the preservation of trade secrets and confidential information. The explanation notes convey that the aim is to strike a balance between ‘the claimant’s rights and the need to ensure that such disclosure would be subject to safeguards to protect the legitimate interests of all parties concerned, such as trade secrets or confidential information’. In other words, the goal is to strike a balance between claimant’s rights and the need for safeguards imposed by the court to preserve trade secrets or confidential information. The court will presume that the defendant did not comply with the duty of care if they refuse to disclose the requested information. In this case, the defendant can remedy that and rebut that presumption by providing evidence.


Recital (20) confirms that national courts should have the power to take specific measures to ensure the confidentiality of trade secrets during and after the proceedings in a proportionate manner in respect of balancing interests. Such measures could include restricting access to documents containing trade secrets and access to hearings or documents and transcripts thereof to a limited number of people. However, the courts cannot decide on this without considering the need to ensure the right to an effective remedy, fair trial and potential harm that could occur.


4. Comment


It is commendable that the EU is taking steps to address the information asymmetry between AI systems’ developers and individuals harmed by their creations. The (prospect of the) realisation of liability and compensation can provide an important incentive to AI providers and users to ensure the safe and correct functioning of their systems. Together with the Product Liability Directive, the proposed AI Act and other product safety rules such as the General Product Safety Directive, the European Commission is designing a comprehensive framework addressing the safety of AI systems and liability for damage caused by those systems.


Nevertheless, the AI Liability Directive harbours an important flaw that might have been overlooked:  the AI Liability Directive offers defendants a way to avoid having to disclose evidence. As the proposal currently stands, if the defendant refuses to provide trade secret information about an AI system as evidence, they will be presumed as non-compliant with the duty of care. A defendant might decide it is strategically wiser to simply pay compensation in exchange for non-disclosure. After all, trade secrets are of great economic importance to such enterprises and one of the conditions for legal protection of trade secrets requires continuous efforts to keep information secret. In other words, non-compliance becomes a choice in order to avoid disclosure.


This is at odds with the drive for transparency of high-risk AI systems in the EU AI Act (art. 13). By offering an option to avoid transparency, the AI Liability Directive undermines this requirement for transparency indicated in the AI Act. This creates tension between these two new instruments. The EC could have chosen a clearer stance on transparency and its necessity, by carrying the requirement of transparency from the AI Act through to the AI Liability Directive.


There is an additional disadvantage to the route the EC has chosen. By avoiding the disclosure of the information necessary to establish fault in liability claims, one can avoid any flaws of the AI systems to be disclosed. This might take away any motivation to improve an AI system, as sufficient financial means offer the possibility to keep any shortcomings of the AI system hidden from the public eye. The lack of transparency could thereby lead to disincentivising the development and improvement of AI systems. Ultimately, this might negatively impact innovation and trust in AI.



Wednesday 11 January 2023

EU/Canada free trade and the Irish constitution: Costello v The Government of Ireland and Ors [2022] IESC 44 - Case Comment



Dr John Cotter, Lecturer in Law, Keele University, UK

Photo credit: Guiseppe Milo, via wikicommons media




The EU-Canada Comprehensive Economic and Trade Agreement (CETA), signed on 30 October 2016 following five years of negotiations, was in retrospect concluded at a pivotal moment in the history of EU trade policy. Though trade policy might not have been the most salient issue in the Brexit referendum earlier that year, prominent Brexiters nevertheless sought to make hay from the EU’s torpid progress in concluding trade deals with third countries, arguing that a nimbler post-Brexit UK would be free to conclude and ratify trade agreements at a faster pace. Less than a month later, in November, Donald Trump won the US Presidential election, which would ultimately put paid to the prospect of completion of the Transatlantic Trade and Investment Partnership (TTIP). Of course, those with a reasonable grasp of EU trade law and policy knew that the signature of CETA marked only the end of the beginning; as a mixed agreement (ie both the EU and its Member States were parties), CETA would require ratification by not only Canada and the EU, but also by all EU Member States. Political opposition, as well as anticipated constitutional and other legal challenges, meant that ratification by all Member States might be a drawn-out affair and was by no means certain.


CETA was approved by the European Parliament on 15 February 2017 and ratified by Canada on 16 May 2017. To date, sixteen current EU Member States have notified the European Council of ratification. On 21 September 2017, in accordance with Council Decision 2017/38, most of the agreement was afforded provisional application. However, several key provisions of the agreement were excluded from provisional application. Of relevance to this blogpost, key provisions of Chapter 8 (investment) were excluded; in particular, those relating to the new Investment Court System (ICS). This ICS comprises a permanent arbitration tribunal (the Tribunal) and an Appellate Tribunal. The Tribunal, which will consist of fifteen members appointed by the EU-Canada Joint Committee established to oversee the application of CETA, will hear and determine claims by investors that a party to CETA has breached certain obligations under the agreement which has resulted in financial loss to the investor. Where a claimant investor is successful, the Tribunal may award compensation, with the parties, including of course EU Member States, being required to recognise and comply with any award without delay. However, execution of any award domestically will be governed by the laws concerning execution of judgments or awards in the state in which execution is sought. It is worth mentioning that an application for execution of a Tribunal award would not have to be brought before courts or tribunals of the defendant state; enforcement could be sought in any state which is a party to the Convention on the Settlement of Investment Disputes between States and Nationals of Other States (ICSID).


As expected, CETA attracted legal challenges, with the ICS being an especial bone of contention. The most notable of these was in Opinion 1/17, in which the Court of Justice, sitting as a full Court, upheld the compatibility of CETA with EU law in the face of concerns expressed by Belgium that, among other things, the ICS might be incompatible with the autonomy of the EU legal order. In March 2022, the German Bundesverfassungsgericht upheld the constitutionality of the provisional application of CETA, though the constitutionality of the ICS in Germany remains an open question. On 11 November 2022, a majority of the Supreme Court of Ireland ruled that ratification by Ireland of CETA as matters stand would be unconstitutional.


Facts, arguments, and the Irish constitutional context

The constitutional challenge to CETA in Ireland was brought by Patrick Costello TD, a Green Party member of the Dáil (the lower house of Ireland’s parliament, the Oireachtas). Pursuant to Article 29.5.2˚ of Bunreacht na hÉireann (the Constitution of Ireland), any international agreement which involves a charge on public funds must be approved by the Dáil in order for that agreement to be binding on the State. Mr Costello argued, however, that CETA could not be ratified by the State without an amendment to the Constitution, which would require a referendum. It was submitted on behalf of Mr Costello that CETA involved the otherwise unconstitutional transfer of legislative and juridical power of the State over to institutions established by CETA. As regards legislative power, it was argued that CETA provided rule-making powers to the CETA Joint Committee which amounted to a power to make laws which would be binding in Ireland in contravention of Article 15.2 of the Constitution, which vests sole and exclusive law-making powers for the State in the Oireachtas.

With respect to juridical power, it was contended that the ICS established under Chapter 8 of CETA would be contrary to Article 34.1 of the Constitution, which provides that “[j]ustice shall be administered in courts established by law by judges appointed in the manner provided by this Constitution”. This juridical power within Ireland held exclusively by Irish courts would be infringed because CETA in conjunction with domestic legislative provisions in the Arbitration Act 2010 would require Irish courts to give virtually automatic effect domestically to awards made by CETA Tribunals. Mr Costello was unsuccessful before the High Court, where Butler J took the view that CETA would bind the State as a matter of international law only and that any decisions of the CETA Joint Committee could not be characterised as laws made for the State within the meaning of Article 15.2. Butler J also held that the disputes to be determined by the CETA Tribunals did not constitute the “administration of justice” within the meaning of the Constitution and therefore would not interfere with the powers of the Irish courts in that regard.

Central to this case in every sense is the concept of sovereignty. Ireland’s history and the anxieties of the framers of Bunreacht na hÉireann leap forth from the constitutional text. The preamble refers to “heroic and unremitting struggle [of our fathers] to regain the rightful independence of our Nation”. The very first article asserts the “inalienable, indefeasible, and sovereign right [of the Irish Nation] to choose its own form of Government, to determine its relations with other nations, and to develop its life, political, economic and cultural, in accordance with its own genius and traditions.” Article 5, with reference to the State, declares that Ireland is a “sovereign, independent, democratic state.” These assertions are given mechanical form in various provisions throughout the Constitution which confer the sole and exclusive law-making power for the State on the Oireachtas (Article 15.2) and the administration of justice in courts established by law (Article 34.1). Consistent with this protectionist approach to sovereignty, Article 29 also establishes that Ireland adopts a dualist approach to international law obligations. Those with pre-existing knowledge of Ireland’s history in the EU will be aware of these facts. In order for Ireland to join the then EEC in 1973, which involved the ceding of some sovereignty, and of legislative and judicial power in the State to European institutions, a referendum had to be held in 1972 to approve the requisite amendment to Article 29. In 1986, when the Irish government sought to ratify the Single European Act sans a constitutional amendment, lawyers on behalf of Raymond Crotty in Crotty v An Taoiseach were successful in convincing the Supreme Court that an international agreement involving the cession of external sovereignty would require a further amendment to the Constitution and, therefore, a referendum. Europe has had to hold its breath while awaiting the results of a number of Irish constitutional referendums since. Mr Costello was evidently hoping to add another chapter to that story.


Supreme Court judgments

Hogan J in his judgment opined that the appeal before the Supreme Court “may yet be regarded among the most important which this Court has been required to hear and determine in its almost 100-year history” (para. 9). The appeal also resulted in a deeply divided court, with each of the seven judges authoring substantial (and in some cases lengthy) judgments which require close examination to reveal the ratio of the case. Mercifully, the judges were able to whittle the appeal down to six issues, set out in the judgment of Dunne J (para. 13):

i)                    Whether ratification of CETA was necessitated by the obligations of membership of the EU.

This had been an argument advanced on behalf of the State in the alternative to the assertion that CETA could be ratified utilising normal processes under Article 29. All seven judges (O’Donnell CJ, Dunne, Hogan, Charleton, McMenamin, Power, and Baker JJ) rejected what one suspects was a rather half-hearted argument, ruling that EU membership did not necessitate ratification of CETA.

ii)                   Whether CETA amounted to a breach of Article 15.2 of the Constitution (sole and exclusive law-making power of the Oireachtas).

It had been submitted on behalf of Mr Costello that CETA would involve interference with the law-making powers of the Oireachtas for the State. In this regard, Mr Costello’s lawyers pointed to the jurisdiction of the CETA Tribunals to make awards against the State for losses suffered by an investor as a result of the operation of a provision of Irish law, arguing that the threat of such awards would create a ‘regulatory chill’ which might prey on the minds of Irish law and policy makers. The Supreme Court was divided on this question, with the majority (O’Donnell CJ, Dunne, McMenamin, Power, and Baker JJ) ruling that CETA would not amount to an interference with the law-making powers of the Oireachtas. Hogan and Charleton JJ, dissented on this point, with the former pointing in particular to the fact that CETA provides for a “form of strict liability on the part of the State in respect of legislation which is found to be contrary to CETA and insofar as it does not contain a good faith defence” (para. 14).

iii)                 Whether the creation of a CETA Tribunal amounted to the creation of a parallel jurisdiction or a subtraction from the jurisdiction of the courts in Ireland contrary to Article 34 of the Constitution (conferral of ‘administration of justice’ in courts established under the Constitution).

In essence, the issue amounted to whether one viewed the CETA Tribunal as a body which would hear and determine disputes which were purely matters of international law (not constitutionally problematic) or whether it amounted to a body which would (or could) become involved in disputes that might otherwise have been heard and determined in Irish courts (possibly constitutionally problematic). Again, the Supreme Court was divided on this issue, with the majority (O’Donnell CJ, McMenamin, Power, and Baker JJ) of the view that CETA did not involve the impermissible withdrawal of disputes from the jurisdiction of Irish courts. Dunne, Hogan, and Charleton JJ differed on this point, though the former two judges pointed to the fact that their chief constitutional objection in this regard arose from the fact that the judgment of a CETA Tribunal would be, in Hogan J’s words, “virtually automatically enforceable” in Ireland (para. 15).

iv)                 Whether the ‘automatic enforcement’ of a CETA Tribunal award by virtue of the enforcement provisions of CETA in conjunction with the provisions of the Arbitration Act 2010 is contrary to Article 34 of the Constitution.

As Ireland is a dualist state, an award granted by an international tribunal will not enjoy automatic enforcement in Irish courts unless such enforcement is provided for specifically under Irish law. In the absence of such a domestic law, therefore (unless one takes the view that the primacy of EU law or the principle of sincere cooperation under Article 4(3) TEU would require enforcement of CETA awards domestically), CETA awards would not be enforceable in Ireland. However, sections 24(1) and 25(3) of the Arbitration Act 2010 give force of law in Ireland to the UN Convention on the Recognition and Enforcement of Foreign Arbitral Awards 1958 (the New York Convention) and ICSID respectively. CETA in turn provides that awards made by the CETA Tribunals are awards for the purposes of the aforementioned international law instruments, meaning that following ratification, CETA Tribunal awards would, save in very limited circumstances, be automatically enforceable in Ireland. Automatic enforcement of CETA Tribunal awards would raise a constitutional difficulty in that it would effectively confer the final decision in a dispute concerning the ‘administration of justice’ within the meaning of Article 34.1 to an international tribunal rather than the Irish courts. On this point, a majority of the Supreme Court (Dunne, Hogan, Charleton, and Baker JJ) held that the virtual automatic enforcement of CETA Tribunal awards in the State would be a violation of Article 34. The reasoning of Hogan J was particularly interesting in this regard; borrowing apparently from the jurisprudence of the German Bundesverfassungsgericht, Hogan J found that the combination of CETA and the 2010 Act would mean that the Irish courts “would have no power to refuse enforcement even where the award compromised Irish constitutional identity or constitutional values in a fundamental way or where it was inconsistent with the requirements of EU law” (my emphases).

v)                  Whether the effect of the interpretative role of the CETA Joint Committee and its role are a breach of Article 15.2 of the Constitution.

Article 25 of CETA allows the CETA Joint Committee to make interpretative decisions which are binding upon the CETA Tribunals. A question arose as to whether these interpretative decisions would constitute an interference with the sole and exclusive law-making function for the State of the Oireachtas under Article 15.2. On this issue, the majority of the Supreme Court (O’Donnell CJ, Dunne, McMenamin, and Power JJ) held that that interpretative role of the CETA Joint Committee was constitutionally permissible. Again, the dissents (Hogan, Charleton, and Baker JJ) were noteworthy. Hogan J, in particular, opined that CETA Joint Committee interpretative decisions amount to “a form of quasi-legislation” which in practice would involve a de facto amendment of CETA without the constitutionally mandated prior consent of the Dáil under Article 29.5.2˚ (para. 17).


vi)                 Whether an amendment to the Arbitration Act 2010 to alter the ‘automatic enforcement’ of a CETA Tribunal award would allow ratification of CETA without an amendment to the Constitution and attendant referendum.

The majority of the Supreme Court had held that ratification of CETA would be unconstitutional owing to the fact that CETA Tribunal awards would be virtually automatically enforceable arising from the interaction of CETA and the Arbitration Act 2010, which would constitute an interference with the constitutionally defined jurisdiction of the Irish courts. One might be forgiven therefore for concluding that ratification of CETA would require a constitutional amendment and a referendum. The Supreme Court, however, took it upon itself to signpost a less complicated way out of the quandary. Hogan J suggested that amendments to the Arbitration Act 2010, which would empower the Irish courts to refuse to give effect to a CETA Tribunal award (on the grounds of Irish constitutional identity or obligations under EU law), would cure the unconstitutionality identified by the majority of the Court (paras 228-237). All of the judges of the Court, save for Charleton J agreed that Hogan J’s prescription would cure the unconstitutionality (or be constitutionally permissible, in the case of those judges who saw no unconstitutionality in the first place). Charleton J’s “ultimate dissent” posited the view that the suggested amendments to the 2010 Act would be ineffectual since the primacy of EU law flowing from Ireland’s obligations to the EU under Article 29 of the Constitution would render it impossible to refuse to enforce a CETA Tribunal award on grounds such as Irish constitutional tradition (para. 62).




Viewed formalistically, the Costello case is purely about the Constitution of Ireland, not CETA itself or indeed, in a direct sense anyway, EU law. There is certainly more than enough in the Supreme Court judgments to engage Irish constitutional scholars for another century. However, the judgments of the Supreme Court have much of interest to say to trade and investment lawyers, as well as EU lawyers, and it is on these points that I will focus my observations.


In order to assess constitutionality, the judges had to analyse the CETA agreement and in many cases make assumptions about its likely operation. In examining the possible effects of CETA on Irish legislative and juridical sovereignty, Hogan J, for instance, took what might be described as a precautionary approach in entertaining (perhaps remote) hypotheticals in which CETA Tribunal awards might result in inhibiting the formulation of legislative policy by the Oireachtas. Hogan J also identified the interpretative role of the CETA Joint Committee as potentially quasi-legislative in nature, with the possibility that this role could be used to effect de facto amendments to the text of CETA. Ironically, it may be the historic expansive interpretation of the EU Treaties by the CJEU, some of which have arguably amounted to de facto textual amendments, that have led to some fear about international or supranational institutions using interpretative powers to change the nature of an international agreement beyond what was ratified at national level. O’Donnell CJ, in contrast, seemed to take a much more sanguine approach to how international agreements like CETA operate in practice and to show concern for what an overly cautious and sovereigntist approach might have on the ability of the executive to conclude international agreements. On a related note, while Hogan J’s suggested constitutional cure was accepted by a majority, Charleton J maintained forcefully that the primacy of EU law would nullify the effect of any such amendments. In the perhaps unlikely scenario that it transpires that Charleton J is correct on this point, it could result in a situation where CETA was ratified by Ireland based on a misapprehension on the part of the majority of the Supreme Court on the nature of the CETA enforcement obligations. This is a conceivable prospective mess that might have been avoided by a preliminary reference to Luxembourg.


The stinging criticism by Hogan J (a former Advocate General at the Court of Justice) to the CJEU’s approach to CETA in Opinion 1/17 is also notable. Although keen to point out the differing constitutional contexts, Hogan J identified what he saw as weaknesses in the CJEU’s reasoning. Taking a position more defensive of legislative and juridical sovereignty (or autonomy) Hogan J rejected, for instance, the CJEU’s notion that only repeated awards of damages by the CETA Tribunals could impact EU or national regulatory legislation and policy making. Hogan J pointed also to the fact that the CETA Tribunals could in practice disregard CJEU jurisprudence and that there would be no remedy in EU or national law for such disregard, a matter the judge believed to be a “significant structural weakness” in the drafting of CETA.


Likewise of note in the judgment of Hogan J is the appearance of the idea of Irish constitutional identity and constitutional values. Seemingly drawn from the terminology of Karlsruhe, this is the first time that such phraseology has appeared in an Irish court judgment. It should be noted, however, that the use of the terms is limited to the enforceability of CETA Tribunal awards in the State. There is no suggestion that such concepts could be utilised to justify non-compliance with obligations flowing from EU membership; indeed, Hogan J made it abundantly clear that if ratification of CETA were required by obligations arising from EU membership that would have overridden any other possible constitutional objection.


As a final remark, the Supreme Court judgments were received with some fanfare by opponents of CETA on the day of their publication. While they are of major legal significance, they – in reality – place little in the way of legal obstacles to the ratification of CETA in Ireland.




Wednesday 4 January 2023

REPowerEU: a European fiscal space beyond the pandemic


Rosalba Famà, PhD student in EU law, Bocconi University


Photo credit: ThibaultC, via wikicommons media


The European Union is in the process of adopting a new initiative named REPowerEU to manage the current energy crisis which followed the Russian invasion of Ukraine. On May 2022, the Commission presented a comprehensive REPowerEU strategy which includes a proposal to amend Directives on the promotion of the use of energy from renewable sources, on energy performance of buildings and energy efficiency, as well as an ambitious industrial policy program which aims to break free from all energy dependency. Only the latter will be the object of the current analysis and I will refer to it as the REPowerEU plan. This programme is designed to boost reforms and investments dedicated to diversifying energy supplies and will be embedded in the legal architecture of Next Generation EU (NGEU), which is the European plan for the economic recovery after the Covid-19 pandemic. This blogpost, after depicting the geopolitical context of the adoption of REpowerEU, will explain its objectives, its legal structure and its functioning. It will then conclude that NGEU represents a blueprint for further European initiatives which involve common spending in times of emergency.   


As it is well known, in order to tackle the economic consequences of the pandemic, the EU adopted the NGEU plan which consists of taking up common debt of up to 750 billion euro to distribute funds amongst the requesting Member States. The distribution is in the form of loans and grants based on a formula that favors countries most affected by Covid-19 (see Art 12 of Reg 2021/241 establishing the Recovery and Resilience Facility). This unprecedented and ambitious plan has allowed the EU to avoid a serious economic recession caused by the consequences of the lockdowns. Since then, another unexpected geopolitical event occurred being the Russian invasion of Ukraine.    


Before the outbreak of the conflict in Ukraine, EU Member States relied extensively on Russian energy imports which provided gas and oil, both for industrial and household use. Accordingly, the Russian energy supply played a crucial role within the European economy. It became self-evident that the EU’s dependence on Russian oil and gas markets represents a threat for the continuation of the EU’s economic recovery following the pandemic and makes an enfranchisement strategy indispensable. Therefore, with the Versailles Declaration of 10 and 11 March 2022, the Heads of States and Governments of the EU agreed on the need for a plan to phase out the dependency on Russian fossil fuel imports.


This ambitious objective requires massive investment in strategic energy infrastructure having an estimated cost of up to 210 billion euro. As such, the EU is confronted with the need to provide additional large amounts of funding to cover the cost of these new investments. The Recovery and Resilience Facility (RRF), the temporary fiscal capacity dedicated to the pandemic introduced with NGEU, is regarded as a “well-suited” instrument to contribute to the Union’s response to these new challenges. To foster the independence and security of the Union’s energy supply, REpowerEU will channel additional funds to the RRF and allocate these amounts to the Member States to modernize the EU energy infrastructure.


Moving to the legal structure of REpowerEU, the plan will have the form of a Regulation of the Council and the European Parliament which amends Regulation (EU) 2021/241 establishing the Recovery and Resilience Facility (RRF). REPowerEU introduces within the framework of NGEU, an additional chapter of reforms and investments, dedicated to diversifying energy supplies and increasing energy efficiency of the Member States. REPowerEU also amends Regulation (EU) 2021/1060 which lays down Common provisions on European funds, Regulation (EU) 2021/2115 that regulates the European Agricultural Guarantee Fund (EAGF) and the European Agricultural Fund for Rural Development (EAFRD), Directive 2003/87/EC establishing a scheme for greenhouse gas emission allowance trading within the Community, and Decision (EU) 2015/1814 on the Market Stability Reserve for the Union greenhouse gas. Amending certain provisions concerning European funds is crucial to channel additional resources to a new “basket” of the RRF, which will be specifically dedicated to finance energy related targets also called “REPowerEU objectives”.


The legal basis for the RRF is Article 175 (3) TFEU, a provision falling within the scope of the cohesion policy that allows the establishment of specific actions outside pre-existing funds when they are proved to be necessary. This provision is drafted in very broad terms and actions to be adopted only need to comply with a necessity test. The general objective of the RRF stated in Article 4 of Regulation (EU) 2021/241 is to promote the Union’s economic, social and territorial cohesion by mitigating the social and economic consequences of the Covid-19 crisis. REPowerEU amendments aim at increasing the resilience of the Union energy system in two ways, on the one hand through a decrease of dependence on fossil fuels and, on the other hand, through a diversification of energy supplies at Union level.  


At the heart of the REpowerEU implementation is the pooling of pre-existing European resources for new energy related targets and milestones, such as improving energy facilities to meet immediate security of supply needs, boosting energy efficiency in buildings and critical energy infrastructure, increasing the production of renewable energy, addressing internal and cross-border energy transmission bottlenecks and supporting zero-emission transport. To accompany those changes, the workforce must also be requalified towards green skills. The provisional agreement on the Commission proposal reached between the European Parliament and the Council in December 2022 set up that at least 30% of resources distributed under REPowerEU must be allocated to measures having a cross-border or multi-country dimension or effect. Moreover, the co-legislators agreed on establishing measures to tackle energy poverty and to support vulnerable households, SMEs and micro-enterprises particularly affected by energy price increases.   


The REPowerEU objectives will be financed in three different ways. Firstly, by revenues coming from the Emission Trading System; secondly by redirecting unspent NGEU’s loans; and thirdly by channeling pre-allocated resources from shared management programmes. As to the new revenue, additional grants of up to 20 billion Euro will be made available. This amount comes from the auctioning of the Emission Trading System allowances. The amendments to Directive 2003/87/EC authorize the auctioning of the allowances for the period until 31 December 2026 and until the total amount of revenue has reached 20 billion Euro. This revenue will be available to the Recovery and Resilience Facility. Resources from the ETS, disbursed in the form of non-repayable support, represent an example of solidarity and economic redistribution within the EU that favors Member States most negatively affected by the energy crisis. To this purpose, the distribution criteria will consider their energy dependency rate and the share of fossil fuels in gross inland energy consumption.


As to the REPowerEU loans’ component, REPowerEU will allow the allocation of unspent NGEU’s loans to the new REPowerEU objectives. As known to date, not all 27 Member States requested NGEU’s loans within their National Recovery and Resilience Plans. For example, Germany’s recovery plan foresees only the request for grants. Notwithstanding, according to Article 14 of the Regulation establishing the RRF, those Member States may ask for support in the form of loans until August 2023. Indeed, those States have access to finance in the financial markets at very low interest rates because of their high credit ratings. As it is self-evident, they have no appetite for this type of financial support, which is provided by the EU only upon the achievement of ambitious targets and milestones set in the National Recovery and Resilience Plan. This explains why there are currently unspent loans within NGEU. Therefore, REPowerEU aims to mobilize these resources to those Member States that have the need due to their lower financial ratings. To this aim, Member States will have a limited period of time after the entering into force of REPowerEU to communicate to the Commission whether they intend to request additional loan support. Afterwards, the Commission must present an overview of the willingness expressed by the Member States and propose a plan for the distribution of the available resources.


The REPowerEU objectives will also be financed by a percentage of resources coming from other funds established by the Union standard budget, i.e. non NGEU-related. In other words, the REPowerEU will allow the redirection of pre-allocated funds to support new energy targets. This third way of financing the REPowerEU objectives is very innovative as it allows Member States to request to transfer resources under shared management programmes to the RRF. Those transfers consist in up to 12.5% of the national allocation under the Common Provisions Regulation (EU) 2021/1060, up to 12.5% of the national allocation under the European Agricultural Fund for Social Development. The Council has proposed the transfer to the RRF of all or part of a provisional allocation from the Brexit Adjustment Reserve. In this way, requesting Member States can pool European resources already allocated to them under different European spending programmes towards energy related targets. This exercise requires strategically orienting public expenditure in times of emergency.


The procedure for approving revised Recovery and Resilience Plans comprises different stages. First, Member States willing to receive additional funding to reach REpowerEU objectives are requested to submit an updated version of their National Recovery and Resilience Plan including a new REpowerEU chapter. Within the framework of the 2022 Country Specific Recommendations, the Commission has indicated actions needed to face the current energy crisis per each Member State. Any Recovery and Resilience Plan which includes a REPowerEU chapter must indicate reforms and investments energy related to be funded by the RRF, with corresponding new milestones and targets in compliance with the Country Specific Recommendations. It may also contain the scale-up of reforms and investments adopted in the previous version of the Recovery and Resilience Plan. Second, the Commission assesses the plans considering whether proposed reforms and investments are capable of contributing towards the diversification of Union’s energy supply and reduction of dependence on fossil fuels before 2030. Third, in the event of a positive assessment on a proposal from the Commission, the Council approves the Recovery and Resilience Plan by means of an implementing decision setting out reforms and investments to be enforced by the Member States and the financial contribution (see Article 19 of Reg 2021/241).

Although NGEU was built as a one-off operation, the current energy crisis is showing that it is flexible enough to accommodate new emerging needs. In particular, the RRF being based on an ordinary cohesion legal basis seems capable to survive well beyond the pandemic. As long as economic resources are channeled to the RRF, this new facility can accelerate investments and address common strategic objectives. The adoption of a plan such as REPowerEU proves that NGEU is a blueprint for further action as the functioning of the RRF is capable to be used again and again. It also shows the necessity for the EU to equip itself with tools to quickly react to unexpected shocks. It remains to be seen if NGEU will account to a mere temporary fiscal space, confined to Covid-19 and the current energy crisis, or rather a much needed permanent one to boost common strategic investments beyond the pandemic.