O Tell Me the Truth About the Transfer of Preliminary References to the General Court. Three Narratives to Be Questioned, Two Thorny Issues to Be Introduced and One Scenario to be Addressed as the Main Legacy of the Recent Reform of the CJEU Statute

Table of Contents: 1. Introduction. – 2. Some narratives to be questioned. – 2.1. Dealing with oxymoron: the (alleged) specialisation of the General Court. – 2.2. Reducing (or changing) the workload of the Court of Justice? – 2.3. The constitutionalisation of the Court of Justice (and the multiple meanings thereof). – 3. The main legacy of the reform: fostering the verticalisation of the EU judiciary. – 4. Two thorny issues to be discussed. – 4.1. The legislative role of the Court of Justice: time for a reconsideration? – 4.2. How to assess whether the reform reaches its targets?

Abstract: The article discusses the main narratives surrounding the reform, namely the increasing specialisation of the General Court, the reduction of the workload of the Court of Justice and its progressive constitutionalisation, highlighting the seed of truth that can be found in them as well as the misunderstandings on which they are based. Subsequently, it proposes a further interpretation on the long-term impact of this reform, arguing that its main legacy should more precisely be identified as the verticalisation of the EU judiciary and discussing its benefits and risks. Finally, it presents a couple of thorny issues that have not thus far attracted the attention of the debates, namely the role of the Court of Justice in the legislative process regarding the amendments of its own Statute and the benchmarks for assessing whether the reform reaches its targets.

Keywords: Court of Justice – preliminary reference – EU constitutional law – national courts – accountability – institutional balance.

1.   Introduction

Let us say it bluntly: at least in the short term, the transferral of preliminary rulings will not upset the EU system of judicial protection, nor the settled case-law of the Court of Justice of the European Union (CJEU). This reform has not been extensively debated outside the limited realm of EU scholars and many national judges are not aware of it, also because just a few operate in the policy fields transferred to the General Court (GC) and just a handful of requests are raised in those areas from some Member States.[1]

After all, this should not be surprising. Being unaware of this reform means, to some extent, catching a crucial part of its spirit: to change everything, for some things to remain the same. The Rubicon of the transferral of preliminary rulings has finally been crossed, for the sake of closing the chapter opened in 2015, by doubling the members of the GC.[2] Yet this has been done only under specific conditions, aimed at keeping the role of the Court of Justice (CJ) and of the preliminary ruling procedure as close as possible to the existing one.

However, as taught by the Gattopardo, even those who embrace change to keep their role in (r)evolutionary times must accept to undergo some mutations themselves.

Therefore, what transformation will the EU judiciary face after this reform? How will the institution that has probably played the most striking role ever for the process of European integration, at least from a purely legal perspective, evolve in the future?

Standing on the shoulders of the previous literature,[3] this article aims at questioning the foundations of the current debates on the reform and proposing new issues to be discussed as well as a different interpretation on the direction taken by the EU judiciary. The reader should be aware that this article has no truth to tell and no answer to give to many of the issues that it raises. Conversely, it aims to look for them, deepening the debates on a reform that has certainly marked the beginning of a new era for the EU system of judicial protection.

To this end, the article starts by discussing the main narratives surrounding the reform, namely the increasing specialisation of the GC, the reduction of the workload of the CJ and its progressive constitutionalisation, highlighting the seed of truth that can be found in them as well as the misunderstandings on which they are based (2).

Subsequently, it proposes a further interpretation on the long-term impact of this reform, arguing that its main legacy should more precisely be identified as the verticalisation of the EU judiciary. While this evolution may certainly bring some benefits to the EU system of judicial protection, it clearly also entails some risks (3).

Finally, it presents a couple of thorny issues that have not thus far attracted the attention of the debates, but which deserve to be discussed for the purpose of assessing the future of the EU judiciary: the role of the CJ in the legislative process regarding the amendments of its own Statute and the benchmarks for assessing whether the reform reaches its targets (4).

2. Some narratives to be questioned

2.1.  Dealing with oxymoron: the (alleged) specialisation of the General Court

The first narrative that deserves to be questioned in order to understand the reform of the CJEU Statute is the one related to the alleged specialisation of the GC, commonly accepted (albeit more or less strongly debated)[4] in the legal literature as a clear output of the reform. A particularly fervent plea for this argument was also made by President Lenaerts in the hearing of 9 May 2023, before the European Parliament. On that occasion, the President of the CJ reassured MEPs about the ‘ordinary’ nature of the reform, emphasising that having a specialised jurisdiction in very technical matters is typical of many national judicial systems and that the CJEU’s request, therefore, was simply aimed at making the GC a ‘spezialisiertes Gericht der Union, im Dialog mit den Vorlagegerichten der Mitgliedstaaten’.[5]

Certainly, the judges of Bâtiment Thémis will now add to their judicial portfolio a new competence to hear and determine preliminary references in some very circumscribed and specialised areas of law – quite a paradox, for an institution called the General Court.[6] After having experienced in 2019 the creation of specialised chambers in staff and intellectual property cases (something quite unique, looking from 1989 onwards),[7] the GC will now further increase its competences, becoming – at least in the less relevant cases – the interlocutor of national judges in another six very technical subjects.

However, both for procedural and institutional reasons, the argument of the progressive specialisation of the GC is not fully convincing.

Procedurally speaking, the transfer of preliminary rulings will increase the legal areas and the judicial procedures that the GC is called upon to handle. In fact, the policy fields in which the GC now delivers preliminary rulings were rarely dealt with in the past through direct actions. Moreover, the reform expands the range of judicial procedures that the GC is called upon to apply, as it adds the preliminary ruling procedure to the one established for managing direct actions.

Nor can it be said that this reform leads to a further internal specialisation of the GC, i.e. of its chambers. The newly established chamber dedicated to preliminary rulings is actually a meta-chamber, i.e.a chamber composed of 10 judges who simultaneously belong to the previous chambers and to the new one.[8] Therefore, the judges dealing with preliminary rulings will become even less specialised than the others. They will be called on to apply a further judicial procedure, and they will be exposed to a much broader range of legal issues. Moreover, if they act as Advocate Generals for preliminary rulings and as judges in direct actions, they will also be required to play a further role.

Even more shadows appear when this narrative is analysed from an institutional perspective.

First, even though the concept of ‘specialised judicial protection’ can certainly not be framed within a single definition or an established set of features,[9] it should be highlighted that both at national level (as in the case of the German Bundespatentgerichten, just to mention one example) and at EU level (e.g. the Unified Patent Court and the Boards of Appeal of several EU agencies) far greater examples of specialisation can be found.[10] Creating judicial panels composed of both legal and technical experts, assessing technical knowledge during the selection of new judges, offering ongoing training to keep judges up to date on the most innovative technical aspects of a certain policy field… these are all features that do belong to the European idea of ‘specialised judicial body’ and clearly cannot (and probably never will) be applied to the GC. This alone should lead to a more cautious application of the concept of ‘specialisation’ to the GC.

But, most importantly, how can this reform be said to be aimed at specialising the GC, if national courts have no guarantees of being able to dialogue with the ‘European specialised judge’ of those specific policy fields?

As is well known, the allocation of preliminary references to the GC, even in the areas theoretically transferred to its competence, will depend on the interpretation of Article 50b(1) and (2) of the CJEU Statute made by the CJ (alone, since there is no room for the referring judge, the GC or the parties of the national proceeding to have their voices heard). There is no doubt that the CJ will be measured and prudent in interpreting the concepts of ‘independence of the question’ and of ‘coming exclusively within one or several of the transferred areas’ (after all, the reduction of its own workload is at stake!). And it is equally evident that a careful interpretation of the legislation currently in force could already allow a direct dialogue between national courts and the GC.

However, if the reform had really been aimed at pursuing the specialisation of the GC and granting to the latter the ‘fan base’[11] of national judges already evoked in the legal debates, it would not have interposed a guichet unique – and a guichet unique thus composed – between the national judges and the ‘EU specialised judge’. National courts should have been left free to consult and send the question to the EU court they deemed best qualified to answer their doubts. In order to prevent abuses, beyond the safeguards already provided by Article 256(3) TFEU, an internal reallocation mechanism could have been envisaged, modelled on the one that already exists for direct actions.[12] Alternatively, if the drafters of the reform could not refrain themselves from following the vogue of this era, namely that of offering a one-stop shop (an approach which is usually dearer to the business world than to that of judicial remedies, but so be it), the guichet unique could at least be composed equally of members of the CJ and of the GC.[13] If that had been the case, national courts would have felt that they had a specialised, European, counterpart coming to their support in the most technical cases. Conversely, the current legal framework may well give the impression that national courts simply activate a procedure aimed at generating a principle of law that they are subsequently obliged to follow.[14]

All these elements should suggest that greater caution is needed when speaking of a reform aimed at specialising the GC. Certainly, the latter will become, in the future, the manager of a plurality of specialised litigations. But will this make the GC a specialised judge? Or, rather, will it make the latter a ‘harmoniser’, i.e.a court aimed at avoiding the fragmentation of EU case-law? The second scenario actually seems far more likely. And the idea, shared virtually by all scholars, that this reform will only be the first step towards a delegation of competences to hear preliminary references in even more areas further supports this reading.

2.2.  Reducing (or changing) the workload of the Court of Justice?

A further narrative that deserves to be discussed in order to assess the rationale (and the future impact) of the reform is the one related to the workload of the CJ, which this reform is expected to reduce. Intuitively, this argument has a strong self-explanatory basis, given by the sheer fact of the transfer of cases from the CJ to the GC. Moreover, supporters of this argument rightly maintain that it needs to be put into perspective: indeed, even if the decrease in workload were of little significance in the short term, it could still give greater relief in the longer run, when further areas will be transferred to the GC.

However reasonable this teleological key of interpretation may seem, an assessment of the immediate gain of transferring the ‘jewel in the crown’[15] should nevertheless be made. Indeed, during the negotiations, the CJ itself associated the need to reduce its workload with what is most delicate and valuable today: the possibility to dedicate more time and energy to dialogue with national supreme courts and thus gain a better understanding of the peculiarity of each national legal order.[16] And given such an important and delicate goal, which must be pursued without delay, the quantitative impact of the reform needs to be carefully assessed.

This issue has already been addressed by scholars[17] who are familiar with the CJ’s internal working methods (it is a pity that it has not been addressed by the CJ itself in its request[18] and, above all, by the EU legislators, who should have carried out an impact assessment themselves) and this article does not claim to add much to what has already been said.

As for today, it is not very clear how much work the CJ will really be relieved of. However, what appears absolutely evident is that the issue of workload should be looked at from another perspective: what was needed was not to reduce the CJ’s workload, but rather to increase that of the GC, following the doubling of its members.[19]

Be that as it may, the quantitative analysis has to start from what the CJ itself stated in its request for amendment, i.e. that the transferral would hand over 20 per cent of the requests for a preliminary ruling to the GC.[20] Given that, according to official statistics, preliminary rulings account for about 62 per cent of the CJ’s workload,[21] this means that the relief should theoretically amount to 12.5 per cent of the overall cases. However, for the reasons discussed above, this share will not be transferred entirely, since some cases will remain within the jurisdiction of the CJ. Though it is early to draw any conclusions, these first 7 months of application of the new Statute show that, out of 32 references for preliminary rulings related (also) to subjects listed in Article 50b, 29 cases were actually assigned to the GC and 3 instead remained within the province of the  CJ.[22] This brings the relief to around 11 per cent of the overall caseload.

However, if some cases are indeed handed over to the GC, more tasks will be acquired by the same CJ: the preliminary control of the guichet unique of all the references raised by national judges, potentially also involving a Réunion Général if the matter is sensitive;[23] the First Advocate General’s examination of all the rulings made by the GC to check whether the réexamen procedure needs to be activated; and the réexamen itself, even though it will hopefully take place just in a few cases. Mathematically speaking, therefore, very little of that 11 per cent of hypothetical relief is likely to remain.

From a quantitative point of view, therefore, the workload of the CJ does not seem about to decrease (at least to any considerable degree). Yet, it will certainly change in its nature, as the CJ will acquire many more tasks of sheer management and supervision of the work of the whole EU judiciary and, in particular, the GC.

2.3. The constitutionalisation of the Court of Justice (and the multiple meanings thereof)

This brings the analysis to the ‘repeated mantra’[24] within the current literature, according to which the CJ, thanks to the reform, will make a further step towards constitutionalisation.

Straight after the publication of the request (which, as is well known, was very similar to the legislation eventually adopted), Petrić highlighted that ‘the reform […] will likely redefine the roles of the CJ and the GC and push them towards their ideal types: the former towards an EU constitutional court and the latter towards an EU supreme court/council of state’.[25] In the same period, Dermine, arguing on the basis of Krenn’s theory,[26] stated that ‘the reform would certainly further the metamorphosis of the Court into a constitutional jurisdiction […] confirm[ing] the Court’s Luhmannian turn, […] thereby instilling another “dose of Habermas”’.[27] After its entry into force, Iglesias Sánchez pointed out that ‘the Reform Regulation is not as innocuous for the EU judicial structure as it may have first appeared. It partially consolidates the role of the CJ as an EU Constitutional Court’,[28] given that the reform entails ‘an explicit endorsement of the constitutionalisation of the CJ’.[29] On the same line, Sarmiento noted that the reform puts the Court ‘on the road to a constitutional court of the European Union’.[30] Similar views are also echoed by insiders. According to Orzan, ‘a process of constitutionalisation of the CJ seems to be increasingly emerging’,[31] both in functional and structural terms. In the words of Tovo, this reform further constitutionalises the role of the CJ, despite pursuing this goal without a real ‘constitutional moment’.[32]

There is no doubt that the CJ, also thanks to this reform, will be able to focus on the most relevant issues brought to its attention. Moreover, according to recital n 4 of the Regulation amending the CJEU Statute, ‘the Court of Justice is increasingly required […] to rule on matters of a constitutional nature’ (emphasis added). This provision should not be underestimated, being the first time in which the EU legislator has associated the term ‘constitutional’ with the CJ, albeit only indirectly.

Yet now that the dust is settling and the reform has started being applied, the time may be ripe for a deeper consideration of how the role of the CJ will evolve in the years to come.

On the one hand, this reform does not grant the CJ any new powers or functions, since it simply gives the latter the option of transferring some cases (which were in any case within its competence!) to the GC. Moreover, the CJ does not acquire, with this reform, the powers which are typical of many constitutional courts (such as, just to give some examples, annulling the legislation of the federated entities or ruling on the validity of the election of the members of the parliamentary assembly). They would really be the game changer for the constitutionalisation of the CJ (and of the EU legal order as a whole). Yet, they have absolutely not been touched by this reform.

On the other hand, the richness of meanings of the term ‘constitutional’ (leaving aside the semantics: the multiple ways in which the role of the constitutional courts has been implemented in temporal and geographic terms) should rather call for a definition of the type of constitutionalisation that is taking place. Answering this question would require a study that cannot fit in these pages; besides, another contribution to this Special Section already tackles similar issues.[33] Two points nevertheless need to be made in order to provide a general framework for discussion.

First, the type of ‘constitutionalisation’ (if such a term can be used at all) the CJ is evolving towards would certainly be atypical and (unsurprisingly) different from that of the national experiences.

It is undoubtedly different to that – to varying degrees Kelsenian[34] – which can be found in most EU Member States,[35] where a constitutional court is such because its role is to ensure, precisely, that legislative acts are compatible with the principles enshrined in the fundamental law of the State and because it accepts that its primacy over the supreme civil, criminal and administrative courts depends only on the degree of importance and abstractness of the rules it is called upon to enforce. Indeed, after the reform the CJ will continue to exercise a role that in the systems of almost all Member States is left up to a Court of Cassation (for instance, in the field of consumer protection)[36] or to a Supreme Administrative Court (as in the interpretation of the public procurement directives, just to mention one example).

At the same time, to speak of a constitutionalisation à la américaine would also be inaccurate. The filter system envisaged by the guichet unique (and by the filter on appeals, provided for by Article 58a of CJEU Statute, further extended by this reform) and the certiorari of the US Supreme Court have certainly brought these two courts closer, as highlighted by several authors.[37] However, the legal systems remain too distant and the type of review that takes place at EU level (certainly not diffused, but evidently atypical and to some extent centralised through the preliminary reference) cannot be compared to the US one.[38]

Second, and probably most importantly, the type of constitutionalisation allegedly activated by this reform does not correspond to the notion set forth by the Treaty of Nice and subsequently confirmed by the Treaty of Lisbon.

At that time, the idea, though incompletely expressed, did indeed have a clear direction. The CJ was to progress towards a ‘constitutional’ dimension, thanks also to the fact that the GC was expected to become, at the same time, a Supreme Administrative Court: a general judge for adjudicating upon the increasingly technical matters that characterised EC and then EU law; a de facto judge of last instance for all those policy fields in which specialised courts were to be set up according to Article 220 TEC, in respect of which the GC would hear both appeals and preliminary rulings.[39]

A progressive technicalisation of EU law has certainly taken place. The specialised courts, as we know, have had no luck,[40] but EU agencies’ Boards of Appeal have proliferated in their stead.[41] These latter bodies, despite their dubious independence,[42] are in fact the first instance reviewers of many technical policy fields. Their decisions can be appealed before the GC and, also thanks to the extension of the filter mechanism mentioned above, are unlikely to be brought to the attention of the CJ.[43] However, this reform has neither regulated the phenomenon of EU agencies’ Boards of Appeal – about which much should be said and even more should be done[44] – nor has it attributed preliminary jurisdiction to the GC in the areas where national courts call for an increase in the quality of judicial law-making,[45] in which the GC already had a consolidated experience (e.g. trademarks and designs, competition, State aid), and in which it already exercises the role of de facto last-instance court (i.e.in the policy fields covered by the above-mentioned Boards of Appeal).

In a nutshell, this reform has made the GC an equerry of the CJ, called on to relieve the latter of less important cases without, however, acquiring its own distinct status.[46] The GC has definitely not become that ‘European Council of State’ explicitly evoked by GC President Van der Woude at the FIDE Conference in 2021, precisely taking up the approach envisaged since the Treaty of Nice.[47] Of course, this may only be a first step, and the GC – as a rather widespread sub-narrative maintains[48] – may acquire this status in the future. However, those who support this argument should also admit that the guichet unique thus composed and the procedural rules discussed above, as well as the areas chosen for the transferral, do not constitute a good training ground for becoming a supreme administrative court in the future.

If the concept of ‘constitutionalisation’ is thus misleading, how can one read the change that this reform has produced in the role of the CJ?

3.   The main legacy of the reform: fostering the verticalisation of the EU judiciary

The different perspective proposed by this article is that of verticalisation. Indeed, this reform ultimately verticalises EU judicial power. While there is no doubt that this latter concept may well fall within that of ‘constitutionalisation’ (after all, constitutional courts are the supreme courts of the land), speaking of ‘verticalisation’ seems actually more accurate to capture the ultimate legacy of this reform.

First, it highlights the main innovation brought by this reform. A constitutionalisation could have indeed been pursued also by remaining closer to the paths envisaged at the time of the Treaty of Nice: yet, the result would have been far different. Many could argue that the ‘Nice approach’ had already faded away many years earlier and, therefore, that the reform has brought no real innovation. To some extent, they would be right, since unmistakable signals thereof had already appeared with the end of the specialised tribunals and the doubling of the members of the GC.[49] However, a different transferral of preliminary rulings (in different areas, under different procedural arrangements) could have changed the current scenario, keeping it closer to the original intentions. Therefore, the reform has brought the considerable innovation of stating that this judicial setting is the final one (or, more precisely, it is the one that will shape the future of the EU judiciary) and that the path now taken will no longer cross the one previously envisaged.

Second, the concept of ‘verticalisation’ captures the deepest implications of the reform, also on the internal allocation of powers within the CJ.

However silently passed over, the guichet unique is managed by the ‘executive’ of the CJ, i.e. the President, the Vice-President and the First Advocate General.[50] The President of the CJ already exercises important prerogatives such as the power to represent both EU courts externally (vis-à-vis Member States, vis-à-vis national courts),[51] the power to appoint the judge-rapporteur[52] (which obviously brings some influence as to how a certain jurisprudence should or should not evolve),[53] the de facto power to sit in any chamber and replace one of the judges formally assigned to it, and the power to initiate the acts of the Committee 255 for the selection of CJEU new members.[54] To those powers, the reform now adds the one to channel preliminary references and thus steer the development of EU case law towards an innovative direction (by assigning cases to the CJ) or a conservative direction (by devolving them to the GC).

Finally, the concept of ‘verticalisation’ gives the possibility to bring a different perspective into the debates on the impact of the reform of the EU judiciary, currently monopolised by either an explicit or implicit comparison with national experiences. What if the reform of the EU judiciary, rather than being inspired by national legal theories, simply replicates typical approaches of the EU institutional machinery?

Somewhat provokingly, one could argue that, while having very few precedents (if it has any at all) at national level, the relationship between the CJ and the GC is not entirely unknown to other EU institutions.

For instance, seen from the perspective of the réexamen procedure, it replicates – with the obvious caveats given by the different contexts, functions and organisation – the relationship between the Council and Coreper. The matters in which the GC considers that a case requires a decision of principle likely to affect the unity or consistency of Union law, and thus refers the case to the CJ under Article 256(3)(2) TFEU, look similar to the issues listed as ‘B items’ in the agenda of the Council – those on which Coreper could not decide upon and thus brings to the attention of the Council.[55] Conversely, the rulings in which the GC replies to a preliminary reference raised by a national court resemble the issues listed as ‘A items’. Indeed, even if those decisions have already been decided by Coreper (the GC), the Council (the CJ) nevertheless has at least a theoretic possibility of checking them one by one[56] and, where deemed necessary, making amendments accordingly through the réexamen procedure.

What, then, are the benefits and drawbacks of the vertical approach chosen by this reform?

On the positive side clearly stand everything that can be reconciled with a conservative approach: the smoothness in the transition towards the new system (a GC entitled to deliver preliminary rulings in sensitive areas might well have needed a certain period of apprenticeship); the lower or almost zero risk of conflicting case-law[57] (both because the areas transferred have a consolidated case-law to be applied and, let us be honest, because the main ambition of even more GC judges will become to be promoted to the CJ, in order to do the same job on more interesting cases); the confidence given by having a strong and united judiciary to foster the evolution of a legal order whose primacy – it should never be forgotten – does not rest upon an explicit legal provision of constitutional value, but on the case-law of the same EU Courts.

Drawbacks, however, cannot be underestimated as well.

Many have already been discussed elsewhere and will be mentioned here just very briefly. First and foremost, the réexamen procedure: of course, its existence is set forth by the Treaties and thus cannot be challenged as such nor be used as a basis for a critique of the reform. Why, however, was not a single line dedicated, throughout the whole legislative process, to discussing the opportunity to amend the réexamen procedure itself? Pros and cons of such an amendment have already been discussed elsewhere[58] and will not be further addressed here. Yet it is quite astonishing that such a delicate provision has never been discussed publicly, considering that amending this procedure could have completely changed the spirit of the reform, by strengthening the role of the GC, or of the referring judge, or of the parties, or reducing the workload of the CJ (and thus increasing the attention given to the most important cases).

A second drawback, as already extensively debated,[59] is a possible mistrust by national judges in their new interlocutor, if the GC is perceived as a second-class judge. As already discussed above, the CJ will have to be very careful in interpreting Article 50b(1) and (2) of its new Statute, so as to give national courts what they actually want: meaningful and timely answers to their questions, with a reasonable degree of predictability (and, possibly, power of influence) as to who will be their interlocutor. The issues at stake are indeed of crucial importance: a decrease in referrals and incompliance of the referring judge with the principle stated in the preliminary ruling.

Finally, an important risk tied to the approach chosen by this reform, one which has not been debated so far, is the intertwinement between the verticalisation of the EU judiciary and that of national ones.

National systems are indeed becoming more and more familiar with mechanisms that verticalise the judicial decision-making process, to strengthen legal certainty and/or shorten the length of proceedings. The saisine pour avis de la Cour de Cassation of the French judicial system[60] or the possibility set forth by Article 363a of the Italian civil procedural code[61] for a first-instance judge to refer a case to the Court of Cassation for a preliminary ruling are good examples, even if certainly not the only ones. The same also applies, under a different perspective, in the relationship between ordinary judges and national constitutional courts: in many EU Member States, constitutional courts are pushing to take on a role of managing the interplay between the national and the EU legal orders so as to increase their control on the protection of fundamental rights, to the detriment of the direct dialogue between national judges and the CJ. It is worth considering some examples: the recent judgment of the Italian Constitutional Court in case No 181/2024,[62] which encourages national judges to refer matters also related to the interpretation of EU law to the same Constitutional Court, as the latter allegedly finds itself in a better position to protect fundamental rights, and could itself potentially enter into dialogue with the CJ;[63] the German Bundesverfassungsgericht’s approach of directly applying the fundamental rights set forth by the EU legal order in cases on the right to be forgotten;[64] the Austrian Constitutional Court’s judgment of 14 March 2012,[65] which led to the A v B case before the CJ;[66] and of course, going even further back in the past, the French mechanism for reviewing the constitutionality of laws, which led to Melki and Abdeli.[67]

Against this backdrop, it seems interesting to evaluate what might arise from the interplay between these two lines of verticalisation. Should the GC not be perceived as a suitable interlocutor, national judges might prefer to dialogue with their own supreme judges. After all, in areas where settled case-law is in place (as it is in those transferred by the reform!), EU law could also be interpreted and applied even without the intervention of the CJ, and the support given by the national Court of Cassation might be perceived as enough. Similarly, in a context where the constitutional court encourages, to say the least, the use of internal mechanisms of review in lieu of preliminary references to the CJ, the perception of dialoguing with a lower tier of the EU judiciary might lead national judges or the parties in national proceedings to take the former route, instead of the EU one. The possibility of avoiding the réexamen and the (at least) one-month period in which the losing party could push for a new reference[68] or for different interpretations might also be appealing. Moreover, applying to national supreme courts, whether ordinary or constitutional, does not preclude the possibility of raising the matter at EU level, since a preliminary reference could well be submitted by the same supreme/constitutional court itself. And many might think that such a reference could hardly be transferred to the GC.

A prudent application of the new provisions could certainly reduce those risks – besides, the GC has already received some cases raised by supreme national courts in the areas transferred.[69] However, this issue should not be underestimated. If the verticalisation of the EU judiciary were to foment (and intertwine with) similar trends at national level, the preliminary ruling procedure could change considerably… while remaining the same; and, with it, EU law itself. As observed even in novels,[70] the dialogue raised by supreme courts is inevitably more frictional, being more easily permeated by national interests. Not coincidentally, the CJ erected the pillars of the EU legal order, thanks also to those whom the former Advocate General Tesauro affectionately called the piccoli giudici.[71]

4. Two thorny issues to be discussed

By way of conclusion, it is worth turning our gaze to a couple of thorny issues that have been somehow obscured by the debates discussed above and that this reform has either shown in all their relevance or revealed to be questions that need urgently to be addressed: the power of the CJ to initiate and steer the legislative amendments of its own Statute and the benchmarks for assessing the output of this reform.

Clearly enough, dealing with these issues means entering into the realm of the CJ and balancing the opposite needs of protecting its independence and keeping it accountable. Thus, a few caveats are worth making.

Article 281 TFEU expressly grants the CJ the power to request amendments to its own Statute. The EU legal system, autonomous and distinct from the national ones, has its own peculiar balance of powers and any reading of the former on the basis of legal principles or theoretical approaches of the latter can only lead to fallacious results. It is typical of any international court – such as, at least since its genesis, the CJEU has been[72] – to have an important say in the management of cases and of the rules of procedure. The participation of the CJ in the legislative procedure leading to the amendment of its own Statute allows the decision-making process to rely on information and experience-based knowledge which is available exclusively to the CJEU; this asymmetry of information embodies and, at the same time, protects the independence of the CJEU itself.[73] Last but not least, if one praises – as is warranted, looking at the recent and past history of the EU integration process – the fundamental role that the CJ has played in shaping the EU legal order,[74] and if the same CJ needs certain structural changes in order to continue exercising this role, then for the sake of coherence it is inevitable to accept a particularly active role played by the latter in stimulating and shaping these changes.

4.1.  The legislative role of the Court of Justice: time for a reconsideration?

Quite apart from all that, this reform brings out a contradiction: does the CJ’s power to initiate the amendments of its own Statute still serve its intended purpose or has it spilled over, becoming a power that inhibits the EU political institutions from building their own political view on the future of the EU judiciary?

Of course, the EU political institutions have expressed their position and exercised their prerogatives throughout the many chapters in which the long process of reform of the EU judiciary has unfolded, from 2015 onwards. Yet, they have often been limited visions, aimed at protecting each institution’s own interests. The Council has certainly shaped this decade of reforms, but mainly for the purpose of protecting the interests of the Member States to be equally represented in the EU Courts – as demonstrated by the failure of specialised tribunals and, above all, by the doubling of the members of the GC. The Commission did not fail to make its position clear, between 2018 and 2019, when the CJ (who else?) had tried the other viable route to an efficient reallocation of the new resources gained after the doubling of the GC, i.e.by proposing the transferral of infringement proceedings.[75] In particular, the Commission immediately stopped the initiative, in order to protect its own secret garden from possibly more lengthy and less efficient procedures.[76] The European Parliament was perhaps the most enterprising institution in this last amendment of the CJEU Statute, as it managed to achieve some changes in the field of participation and transparency.[77] But which EU political institution, in the last decade, has expressed a comprehensive vision of judicial policy? None: each one has intervened to protect its own interests or, as in the case of Parliament, to wage side battles on issues certainly relevant per se (and easy to sell during an electoral campaign) but far from the heart of the debate and in any case incapable of expressing an overall vision.

In this framework, the CJ has emerged as the only institution capable of defining (and implementing) a political strategy, also strengthening its role over time. Leaving aside the quarrels surrounding the doubling of the GC,[78] on that occasion the CJ was the entity that succeeded in finding a compromise solution for solving the existing conflicts. In 2024, it did not limit itself to this role but rather dealt the cards and led the dances from the opening to the closing of the legislative procedure.

Let us put it clearly: here and now, any alternative to this could only produce worse results. But equally bluntly, it needs to be highlighted that this situation reveals a distortion of the system, which should therefore be addressed.

To this end, looking to the future chapters that will be written in the coming years, it seems legitimate to question the degree of intensity of the CJ’s participation in the negotiations – at least in those cases where the discussion does not refer to mere procedural rules, but to aspects of the Statute that shape the overall judicial architecture of the Union.[79]

In particular, it is worth assessing to what extent the power to stimulate an amendment of the Statute – for the caveats stated above, of a fundamental value – might give rise to the possibility of participating in trilogues (rectius, quadrilogues) on a par with the political institutions, and from there drafting the legislative act.

Such participation may well be justified by the general principle of decision-making efficiency, together with that of loyal cooperation among EU institutions. Moreover, as is well known, trialogues are a realm of informality and (perhaps unfortunately) do not have strict provisions on the criteria for participation.

Still, neither the 2007 Joint Declaration on practical arrangements for the co-decision procedure[80] nor the 2016 Interinstitutional Agreement on Better Law-making[81] (i.e. the only kind of ‘legislative framework’ for trilogues)[82] mention the CJ. The same applies for the European Parliament’s internal regulation, which regulates the possibility for this latter institution to participate in trialogues with the Council and the Commission, without mentioning the CJ, and without using general clauses referring, in broader terms, to the institution presenting the legislative proposal[83] (as is well known, the internal regulations of the Commission and Council do not even mention trilogues).[84]

And this brings the discussion to a more striking aspect, namely the fact that, for the purpose of amending its own Statute, the CJ does not adopt a legislative proposal but rather submits a request to amend the Statute. Article 281 TFEU is indeed very clear in providing for two different options for initiating the procedure for an amendment of the CJEU Statute: one is at the request of the CJ, after consultation with the Commission; the other is on a proposal of the Commission, after consultation with the CJ. This difference clearly reflects the different roles of the two institutions. While there is no doubt that, as discussed above, the CJ has very broad powers to define the rules of procedure governing the disputes pending before it, the Treaties have nevertheless put some limits on these prerogatives, since the Court is an expression of a judicial power, not a legislative or political one. A concrete example of these limits is the difference, which cannot be only semantic, between the ‘request for amendment’ and the ‘proposal for amendment’ set forth by Article 281 TFEU. Can, therefore, the rules on trilogues relating to those who make a proposal for amendment also apply, per relationem, to those who merely make a request?

In this regard, it is interesting to note that such an extensive interpretation is not made in relation to other procedures. For instance, in the procedure under Article 7 TEU, the institution initiating the process is not allowed to participate in discussions in the Council as to whether there exists a clear risk of a serious breach of EU values. After having initiated the procedure against Hungary, the Parliament was not admitted to take part to the negotiations in the Council – even if the act adopted to initiate the procedure was, as Article 7 TEU explicitly states, a ‘reasoned proposal’ and even if, being author of such a reasoned document, the Parliament could well have brought important information into the discussion. It goes without saying that the rationale of this is that the procedure set forth by Article 7 TEU is largely intergovernmental and, therefore, a supranational institution such as the European Parliament does not fit in. It has the power to initiate it, but not the right to participate in the deliberations.

But is the CJ not in the same situation when it comes to the ordinary legislative procedure? It has the power to present a request for amendment, yet not to participate in the deliberations—at least, not on a similar footing with political institutions. The possibility set forth in the internal regulations of both the European Parliament and the Council to organise hearings with other institutions[85] could enable them to bring the expertise and the voice of the CJ into the legislative procedure, while leaving the responsibility for judicial policy choices on the shoulders of the EU political institutions.

4.2. How to assess whether the reform reaches its targets?

The second, equally thorny, issue that deserves to be brought into the discussion, also for the sake of further addressing the ‘futurewilltell-ism’ that, as already stigmatised, heavily affects the current debates,[86] regards the identification of concrete benchmarks for assessing whether the reform is able to reach its targets.

The fact that (apropos of lack of political vision) no official document clearly defines those targets (and that even scholars struggle to agree upon them)[87] points to the need for a preliminary assessment of the overall objectives of the reform. To those already highlighted in literature, this article adds some new ones, using as a source the already mentioned hearing that took place on 9 May 2023 before the European Parliament.[88] On this occasion, President Lenaerts identified, as the main objectives of the proposed amendment, the need to i) rebalance the workload between the two EU courts, ii) reduce the length of proceedings, and iii) improve the quality of judicial decisions.[89]

Little needs to be said on the first point, at least from a quantitative perspective: there is no doubt that a reallocation will take place and a better balance between the workloads of the CJ and GC will be found accordingly. Of course, the extent of this reallocation will certainly deserve to be measured, but the task is an easy one.

The other two points are far more interesting.

The length of proceedings should be measured, according to what President Lenaerts stated on that occasion, not in average terms, but taking account of the various procedures. Indeed, between 2016 and 2022 the average length of the preliminary ruling procedure increased by only 2.3 months, from 15 to 17.3 months. However, this figure also includes all the cases decided with expedited procedures, which of course pull down the average. If one looks only at the most important cases, where the protection of fundamental rights is at stake or where the dialogue with national supreme courts is dedicated to issues of a constitutional value, the average stood (as reported by the same President) above 2 years.[90]

This analysis cannot but be fully supported. The experience of preliminary references raised by Italian Courts (the only ones for which such data have been made publicly available by academic research, at least to the author’s knowledge) tells us that in 2016 the average length of the preliminary reference procedures concluding with an ordinary judgment was 20.9 months and was up to 23.4 months by 2022. In the same timeframe, the average duration of all the proceedings (ordinary and expedited) was a far more reasonable 17 months (in 2016) and 18.2 months (in 2022), thanks of course to the impact of the cases decided with orders.[91] If one looks at the impact of the composition of the chamber on the length of the procedure, the Italian experience[92] shows that, from the entry into force of the Lisbon Treaty to 31 December 2024, the average length of the cases decided by ordinary chambers (3 or 5 judges) was about 15 months, while that of those decided in the Grand Chambre was about 22 months.[93]

For the time being, the CJ does not publish such data in its yearly official statistics, where only the overall duration of the procedure is given, and only the impact of urgent and expedited procedures is specified.[94] However, it should start to break up the data related to the length of preliminary rulings, also highlighting the differences at least with regard to the composition of the chamber and between cases decided with a judgment and those decided with an order. More precisely, this latter item could be broken down at least between ‘Cilfit orders’ (i.e.those adopted pursuant to Article 99 or 100 of the CJ Rules of procedure for bringing to the attention of the referring judge that the question referred is identical to a question on which the CJ has already ruled, that the reply to the question raised may be clearly deduced from existing case-law or that the answer to the question referred admits of no reasonable doubt) and the others. Of course, those data should then also be presented in the report which, according to Article 3 of Regulation (EU) 2024/2019, the CJ has to present to the European Parliament, the Council and the Commission by 2 September 2028 for the purpose of assessing the implementation of the reform. This provision (see Article 3(2)(c)) only requires that the average length of the procedures be reported. However, for the reasons stated above, far more precise data are needed to properly assess whether the reform succeeds in reaching the target to reduce the length of the most sensitive cases.

With regard to the need to increase ‘la qualité des décision judiciaires’, it should be highlighted that, according to President Lenaerts, this implies ‘la capacité d’étudier à fond le dossier, d’étudier la jurisprudence nationale des cours constitutionnelles, suprêmes, qui sont une source de nourriture de notre jurisprudence’.[95]

In consideration of this target, one cannot but highly appreciate the commitment taken by the CJ to rely more on Article 101 of its Rules of Procedure and, accordingly, to increase the practice of requesting clarification from national judges in order to better understand the questions referred by the latter and the peculiarity of each national legal order. The number of activations of this provision will be stated in the forthcoming 2028 report (or, at least, so it seems reasonable to infer from Article 3(2)(g) of Regulation (EU) 2024/2019, which should have probably been written more clearly). Yet, the CJ should also consider adding this information in the official statistics published yearly.

On the other hand, the legal basis requesting the 2028 Report does not envisage the possibility of analysing national jurisprudence for the sake of evaluating the relationship between national and EU Courts in the fields where preliminary rulings have been transferred to the GC. This may well create a sensitive lacuna in the evaluation of the reform, which should be addressed under two perspectives.

First, by analysing the follow-up decisions adopted by national judges after having received the answer to their question from either of the two EU Courts, so as to assess whether and to what extent they comply with the principles stated by the latter in the fields covered by the reform and to assess the impact of the author of the preliminary ruling. Second, by looking into national case-law to assess whether, when and why national judges refuse to raise a preliminary reference. For the reasons already discussed above, maintaining a sound, productive relationship with national judges, despite the appearance of a third party in the dialogue, is perhaps the main challenge posed by this reform. Enriching the 2028 Report with an analysis of these two issues, either made by the Recherche et Documentation of the same CJEU or possibly by creating a pool of national experts, would certainly elevate the debates on the future extension of the GC’s competence in preliminary rulings over new policy fields.

Along the same line, it is worth considering to disclose more information with regard to hearings. The provision regarding the 2028 Report does not mention this issue and the yearly official statistics offer very scant information on that. However, since the 2012 amendment of the CJ Rules of Procedure,[96] which gave the CJ the option of refusing the request of the parties to be heard if it considered that it had sufficient information to give a ruling,[97] the organisation of hearings has decreased dramatically. Academic research – official statistics, unfortunately, do not cover this subject – reveals that from 1 November 2012 to 31 December 2023, a hearing was held in only 31 per cent of the preliminary references raised by Italian judges, 30 per cent of those coming from Spain, 32 per cent of those from Germany, and 39 per cent of those from Belgium.[98] For some Member States, statistics are slightly better (47 per cent and 56 per cent for French and Dutch cases, respectively).[99] Yet, also considering that Italy and Germany are in the top two positions for number of references,[100] these data clearly reveal that the 2012 amendments of the Rules of Procedure have enabled the CJ to greatly reduce the oral phase of preliminary reference procedures.

However, for EU Courts, hearings are a further moment in which information on the ever-evolving national case-law can be obtained. Moreover, hearings are of crucial relevance for the parties both to defend their interests and to make them feel that they are being heard and are receiving an answer that is useful for solving the litigation.

Therefore, the resources gained through the partial transferral of preliminary rulings call for a reconsideration of such a restrictive approach of the CJ (and, potentially, of the GC) towards the organisation of hearings in preliminary rulings, as well as for having official data thereon. Indeed, academic research shows that, at least with regard to the preliminary references raised by Italian courts in the timeframe mentioned above, the length of proceedings where a hearing took place is lower than that of those where only the opinion of the Advocate General was presented.[101] This means that, at least in cases where the CJ decides to hear the AG’s opinion, the organisation of a hearing does not extend the length of the preliminary ruling procedure (paradoxically enough, it even reduces it, by only a few weeks though). Thus, the added value that the latter can bring in terms of awareness of the parties’ needs, national case-law, or possible amendments that may have occurred in national legislation or jurisprudence after the closing of the written phase (or possible different interpretations thereof: as is well known, the parties cannot reply to the others’ observations, if there is no hearing) can be obtained without resulting in an extension of the overall length of the procedure.

Therefore, the judicial statistics published each year, as well as the forthcoming 2028 Report, should carefully consider the possibility of providing data on hearings and on their impact on the length of proceedings.

Indeed, it is (also) against those practical benchmarks that it will be possible to assess the impact of this reform and, more importantly, take reasoned decisions on the next steps of the EU judiciary.

-------------------
European Papers, Vol. 10, 2025, No 2, pp. 333-356
ISSN 2499-8249 - doi: 10.15166/2499-8249/835