Abstract: The end of 2022 saw the publication of the proposal for reform of the preliminary ruling procedure, drafted by the Court of Justice of the EU. If the proposal goes through, it will bring about one of the most significant changes in the EU judicial system in its seventy years’ history. In this Insight, I start with an overview of the background to this reform proposal in order to provide the relevant context, before turning to its content. The focus is on specific areas of EU law in which the General Court would acquire jurisdiction in the preliminary ruling procedure. Then I discuss some downsides of the proposed distri-bution of powers between the Court of Justice and the General Court and suggest certain adjustments to the practice of the two courts to make the reformed preliminary ruling mechanism respond better to its objectives. Before concluding, I also reflect on the possible impact of the reform on the most numer-ous members of the EU judicial system – national courts.
Keywords: preliminary ruling procedure – jurisdiction – Court of Justice – General Court – specific areas of EU law – national courts.
December 2022 witnessed a small but significant moment in EU law history. The Court of Justice published a proposal whose first and more important part concerns the reform of the preliminary ruling procedure under art. 267 TFEU. According to the proposal, the General Court would take on answering preliminary references from national courts in several specific areas of EU law. This raises a number of interesting questions for the functioning of this mechanism, which was dubbed “the ‘keystone’ of the Union’s judicial system”, and whose object is “securing uniform interpretation of EU law, thereby serving to ensure its consistency, its full effect and its autonomy as well as, ultimately, the particular nature of the law established by the Treaties”. What seems certain is that after some twenty years of resistance, the Court of Justice appears to be willing to let the General Court take a “jewel” from its “crown” and wear it on special occasions.
In this Insight, I will first introduce the background to the 2022 reform proposal, and then move on to its content. The focus will be on specific areas of EU law in which the General Court should have jurisdiction in the preliminary ruling procedure. In particular, I will discuss what the downsides are in my view of the proposed distribution of powers between the Court of Justice and the General Court and the procedural adjustments that could be adopted through the practice of the two courts to make the reformed preliminary ruling mechanism respond better to its objectives. Last, I will also reflect on what happens to those treated by the reform proposal as a “third wheel”, in other words, the national courts.
II. How did we get here?
The possibility of transferring jurisdiction in the preliminary ruling procedure to the General Court came with the 2003 Treaty of Nice. More specifically, it was provided in art. 225(3) of the then EC Treaty, which is now art. 256(3) TFEU. Although it was discussed extensively then, this possibility was never realised.
Around the same time, the preliminary ruling procedure was coming under ever greater pressure. The EU was about to expand to ten new Member States, increasing the total number of national courts that kept bringing most of the cases before the Court of Justice. With a series of Treaty amendments from the mid-1980s to the late 1990s, EU competences were extended into new regulatory areas, in parallel with the expansion of the Court’s jurisdiction in those areas. The Court’s workload was constantly on the rise, as was the time needed to resolve cases. And the greatest chunk of that load came from handling preliminary references. So, the consensus was that something had to be done.
Enter the General Court, then the Court of First Instance, which was identified as a possible solution to the problem. However, most of the insiders and outsiders in the EU judiciary were cautious about the prospects of delegating certain matters in the preliminary ruling procedure to the General Court in the run up to the Nice amendments and in their aftermath. The main concerns were of a different nature: legal, concerning the problem of defining in general terms matters that would go before the General Court and delineating them from those that would stay before the Court of Justice; practical, concerning the time national courts would have to wait for a final preliminary ruling if appeals against the General Court’s decisions are allowed or in situations in which the General Court decides to refer the issue back to the Court of Justice on account of its significance; and political, concerning the acceptance by national courts of decisions rendered by the General Court instead of the Court of Justice and their willingness to engage in dialogue with the former interlocutor knowing that its rulings may, albeit exceptionally, be subject to appeal before the latter.
For all the uncertainty surrounding these problems, the solutions were to be sought elsewhere. Some suggested that more should be required from national courts, for example by requiring them to propose answers to the referred questions. Others were in favour of adjustments made in the way the Court of Justice deals with the preliminary references, through, for instance, the introduction of a filtering mechanism or greater use of reasoned orders. The transfer of jurisdiction over preliminary rulings to the General Court indeed proved to be a measure of last resort.
The 2015 reform of the General Court, where the number of its judges doubled, and where the Civil Service Tribunal was abolished, brought that measure back on the table. Yet the Court of Justice kept the same position and saw no need for activating the enabling clause under art. 256(3) TFEU. It offered several reasons for its stance: from the fact that the reform of the General Court was still to be completed, which carried its own challenges, to the fact that the Court of Justice itself was starting to handle preliminary references with more efficiency and expediency, i.e. greater numbers in less time. So, the Court concluded that at that moment the time was not right to move on with transferring partial jurisdiction in the preliminary ruling procedure to the General Court.
This changed by 2022. On one hand, the number of preliminary references kept on rising while the Court’s efficiency and expediency declined. On the other hand, the reform of the General Court was fully completed, and that court seemed to be doing well in its new look. The President of the General Court asserted extrajudicially that they were ready to take on more jobs and assume more responsibility. The Court of Justice soon followed. And now, the ball is in the EU legislature’s court.
III. What is the General Court supposed to get?
The Parliament and the Council will have to decide whether to follow the Court’s proposal, by adopting in the ordinary legislative procedure a regulation amending the Statute of the Court of Justice of the EU, which will state that the General Court acquires jurisdiction to deal with preliminary references that exclusively fall into one or several of the following specific areas: value added tax (VAT), excise duties, customs and tariff classification, compensation and assistance to passengers, and greenhouse gas emission allowance trading.
Under the new division of work, the Court of Justice will keep receiving all incoming references. So, the preliminary ruling procedure will remain centralised, for the sake of legal certainty, predictability, and expediency. In other words, the referring courts and parties in the domestic proceedings will know in advance to whom their questions will be directed. If it were otherwise, national courts would have to decide on their own which court gets their questions, which was expressly excluded as an option in the reform proposal. Everything should therefore go to the Court of Justice. Then, the Court will verify the criteria of specificity and exclusivity. If they are cleared, i.e. if the Court finds that a reference indeed exclusively concerns one or several of the specific areas, the case will automatically be forwarded to the General Court and that court will deal with it.
When the General Court is seised of the case, a number of arrangements will be put in place to ensure a proper environment for the preliminary ruling procedure. The idea is to have the General Court coming as close as possible to what the Court of Justice does in this procedure. First, preliminary references will be dealt with by chambers specifically designated for that purpose. Specialised chambers are nothing new to the General Court, since there are already those that deal, for instance, with staff and trademark cases. Second, for every reference the General Court will assign an Advocate General from among the judges of those specialised chambers. This, however, does not mean that for every reference the opinion of an Advocate General will be issued, since that is not the case before the Court of Justice either. Third, the General Court will be able to hear a reference in a flexible chamber formation, whose size could be anywhere between five (default formation) and fifteen judges (Grand Chamber). This formation would be convened for cases of greater importance and/or difficulty.
After the General Court reaches its decision, it should as a rule be final. The possibility of the review of that decision by the Court of Justice should be limited and exceptional, as already provided in art. 256(3) TFEU. In accordance with that provision, the review would only proceed in the case of “a serious risk of the unity or consistency of Union law being affected”. It would go like this: first, the First Advocate General (currently Maciej Szpunar) has one month to scrutinise the ruling of the General Court and decide whether to propose that the Court of Justice proceed with the review. If he does so, then the Court has another month to urgently review it and deliver a final judgment answering the preliminary reference. For the referring judges who stay the proceedings pending before them and who await a reply from Luxembourg, this means that in the worst-case scenario the whole thing could be prolonged for a maximum of two additional months.
Of all the novelties that the reform proposal brings, the crucial issues in my view are how the Court of Justice came to designate specific areas that should be left to the General Court and when a preliminary reference can be considered as exclusively falling within those areas. These issues are taken up in the following section.
IV. When is a matter specific and exclusive?
As already mentioned, the General Court is supposed to become involved in the preliminary ruling procedure only in several specific areas of EU law. As the Court of Justice noted in its proposal, a general transfer of jurisdiction to the General Court would be precluded by the wording of art. 256(3) TFEU, which expressly refers to “specific areas”. So, sending preliminary references from the Court of Justice to the General Court on an ad hoc basis – e.g. those that concern simpler questions of interpretation of EU law that raise no reasonable doubt about their resolution or whose answers are easily found in the existing case law – was apparently not an option.
This raises the question of how to define relevant “area” of law. The Court of Justice was led by four distinct criteria.
First, the area should be “clearly identifiable” and “sufficiently separable from other areas governed by Union law”. Second, that area should not generally give rise to difficult questions of principle that are of fundamental importance for the EU legal order. Third, a developed case law of the Court of Justice should exist to enable the General Court to navigate through that area to prevent inconsistencies and divergences in the case law of the two courts. And fourth, that area should generate a sufficiently high number of preliminary references, so that the transfer of jurisdiction to the General Court could have a meaningful impact on the workload of the Court of Justice.
Based on those criteria, the Court of Justice conducted a statistical analysis of the recent patterns in the preliminary ruling procedure. It then came up with specific areas that tick the boxes on all four criteria: VAT, excise duties, customs and tariff classification, passengers’ rights, and emission allowance trading.
Regarding the first criterion, in the Court’s view, these areas are all “clearly identifiable” and “sufficiently separable” from other areas of EU law. Moreover, they typically concern a smaller number of acts of secondary law; for instance, the VAT Directive, the Common Customs Tariff Regulation, or the Passengers’ Rights Regulation are what the majority (if not all) preliminary references in these areas are about.
Regarding the second criterion, the Court noted that these areas rarely involve important or difficult questions of EU law. In the reference period, only three cases (of 631 in total) were solved by the Grand Chamber, whereas most of them were dealt with by three-judge chambers, and without the opinion of the Advocate General.
Regarding the third criterion, the Court merely noted that it had generated abundant case law in all these areas. This is tied to the fourth criterion, and that is the generation of a significant number of preliminary references in these areas, which jointly account for around one fifth (20 per cent) of all preliminary references received by the Court per year. The Court of Justice therefore concluded that the proposed sectoral transfer of jurisdiction in the preliminary ruling procedure to the General Court is expected to lead to “a significant reduction” of its caseload.
The selection of specific areas that are supposed to go into the hands of the General Court is not without question. Although some areas that made it onto the list, like customs and tariff classification, have been mentioned for a while, others that were considered in the early days of the idea, such as intellectual property and trademark or judicial cooperation in civil matters or public procurement, were left out. The Court of Justice does not explain why other possible candidate areas were not considered or on what counts they failed to qualify, be it severability, importance, case law, or numbers. It will also be interesting to see whether the EU legislature can add to or take away from the proposed list. If not, other areas could certainly be considered in some future revisions if this first step of the reform of the preliminary ruling procedure eventually goes through.
At the same time, it is not fully certain whether all areas that are currently on the list satisfy all the stated criteria. In particular, the fourth criterion – that the area should generate a sufficiently high number of preliminary references – seems to be missing for some areas. Interestingly, the Court of Justice in its proposal merely notes that all areas together account for roughly 20 per cent of all preliminary references on a yearly basis. However, among them, two carry almost the entire load while others stand apart. As can be seen in the Court’s statistics, of 631 total cases in the six years’ reference period, VAT (286 cases or 45.3 per cent) and passengers’ rights (237 cases or 37.6 per cent) generate more than 80 per cent of the workload, whereas the remaining four areas each generate around 5 per cent or less: excise duties (25 cases or 4 per cent), customs (30 cases or 4.7 per cent), tariff classification (32 cases or 5.1 per cent), and emission allowance trading (21 cases or 3.3 per cent).
Perhaps there are other areas of EU law that generate more preliminary references and would hence represent an even greater relief for the Court of Justice. But those areas probably do not square with the other requirements, especially the second criterion of constitutional importance. So, in those areas the Court decided it would prefer to keep exclusive jurisdiction. In that sense, it seems that the first two criteria are weightier than the second two. This might be relevant for future discussions or revisions of the list of areas that might be placed under the General Court’s watch. If an area of EU law in general does not involve hard questions of constitutional importance or raise important vertical issues (fundamental rights, institutional balance, competences, national values, etc.) and does not intersect horizontally with other areas of law, it could be considered a good candidate for joining the list, irrespective of the absolute number of cases it generates or the stage of development of the Court’s case law – and vice versa.
In any event, if the EU legislature accepts the Court’s proposal and leaves the current list of specific areas of law that ought to come under the General Court’s jurisdiction intact, for a while these areas will be dealt with by the latter court. As already explained, the Court of Justice will keep receiving all preliminary references but will only be able to verify whether they indeed exclusively fall within one or several of these specific areas. If they do, the case will automatically be transmitted to the General Court, no matter how important or politically sensitive the case may be. This is unfortunate, given that some questions, albeit exclusively falling within the currently listed areas, raise questions of interpretation of fundamental constitutional importance for the EU legal order. To give a simplified, perhaps even exaggerated, example, think of the Court’s landmark rulings in Van Gend en Loos or Åkerberg Fransson. At first, the former case might have been understood as being about customs and the latter about VAT. Both areas are currently on the list of areas envisaged to come under the jurisdiction of the General Court. Of course, a careful and informed reader will easily identify in these two references elements that take them beyond the exclusive confines of specific areas like customs and VAT, namely the general issue of direct effect of the EU Treaties and the scope of the EU Charter of Fundamental Rights, respectively, both being of outstanding structural importance for the EU legal order.
But for some references, it may be more difficult to identify at the outset whether they conceal similar important questions that do not concern exclusively the listed specific areas of law. In some instances, the full import of a reference may become apparent only in the later stages of judicial proceedings. So, in these cases, the essential question will be how the Court of Justice interprets the “exclusivity” criterion. Both narrow and wide interpretation bring their own problems. If a defensive Court of Justice pursues a narrow interpretation, the General Court will receive fewer references. This means that the underlying rationale of the reform proposal – to take a substantial burden off the Court’s docket – will suffer. Conversely, if a relaxed Court of Justice pursues a wide interpretation, the General Court might receive cases of greater importance that are more suitable for resolution before the Court of Justice. In this scenario, two safeguards may kick in, both provided for in art. 256(3) TFEU. First, the General Court will get to interpret the same “exclusivity” criterion. If it finds that the reference “requires a decision of principle likely to affect the unity or consistency of Union law”, it may return that reference to the Court of Justice. And second, if that does not happen and the Court of Justice is not satisfied with the ruling of the General Court, there is the possibility of exceptional review if the Court finds that “a serious risk” exists that the unity or consistency of EU law would be affected by such a misplaced ruling.
As it becomes apparent, a large part of the success of the reform proposal may depend on the interpretation of the “exclusivity” criterion by the two courts. In my view, how the General Court interprets this criterion will matter more. Why? The General Court will likely have more time and inputs from different sides to delve into all the details of the references it receives. This is very different from the Court of Justice, which should in principle have only limited time to scan the text of the reference and verify whether it falls prima facie exclusively within the listed specific areas or instead overlaps with other areas of law or raises constitutional issues. In the following section, I discuss how the General Court should approach this task, taking into consideration the purpose of the preliminary ruling procedure and the objective of the reform proposal.
V. When should the General Court refer the matter back to the Court of Justice?
The General Court should return to the Court of Justice all references that raise questions of abstract interpretation of EU law and keep for itself all references that involve issues of factual interpretation of EU law. Before moving on to discuss this proposal further, I will briefly explain what I consider by “abstract” and “factual” interpretation.
By abstract interpretation, I understand determination of the meaning of legal provisions. By factual interpretation, I understand determination of which facts come within the scope of the meaning of legal provisions. To give a simple example, which has become a tradition in jurisprudential circles, consider the “No vehicles in the park” provision. Abstract interpretation of this provision would be the determination that it means that no vehicles that may threaten the safety of visitors when operated are allowed in the park, or that no vehicles that pollute when they are operated are allowed in the park, etc. The factual interpretation of this provision would be the determination that a motorcycle while ridden is within the scope of the meaning of this provision, i.e. it is forbidden in the park, or that a bicycle is outside the scope of the meaning of this provision, i.e. it is allowed in the park, etc.
There are several reasons that in my view support the view that factual interpretation should be for the General Court, and abstract interpretation for the Court of Justice. First, art. 256(3) TFEU mentions references that require “a decision of principle”, which the General Court may decide to refer back to the Court of Justice. This “decision of principle” comes closest to what I term abstract interpretation, since in those decisions the Court of Justice would lay down abstract principles that can later be applied to specific factual circumstances. At the same time, the reform proposal itself seems to be hinting that the General Court is particularly suitable for questions of factual interpretation of EU law.
If the General Court indeed rules on the questions of factual interpretation of EU law and refers to the Court of Justice questions of abstract interpretation of EU law, it would still receive a substantial number of references, thus relieving the Court of Justice of some of its significant workload. As the reform proposal noted, specific areas that are meant to come under the jurisdiction of the General Court “raise few issues of principle” and “rarely give rise to judgments of principle”. In those cases, the meaning of EU law is clarified in the case law of the Court of Justice and references are sent to determine whether certain facts are within or outside the scope of that meaning. To illustrate this point, what follows are some examples from the recent case law in the areas included in the reform proposal.
The first example concerns the meaning of the concept of “extraordinary circumstances” under art. 5 of the Passengers’ Rights Regulation, which has been specified by the Court of Justice on numerous occasions. National courts have nonetheless kept referring questions which ask whether unique and ever more peculiar factual scenarios are within the scope of that meaning, such as a passenger biting other passengers and assaulting crew members during a flight.
Similarly, the Court has regularly interpreted exemptions from the application of the VAT Directive provided in its arts 132 and 135, which concern activities such as “school or university education”, the “provision of medical care” or “leasing or letting of immovable property”. National courts have been asking whether the meanings of these exemptions cover activities such as providing driving lessons in driving schools, providing swimming classes in swimming schools, giving medical advice by phone, renting IT equipment cabinets for servers in a computing centre, and so on.
Perhaps the best example of factual interpretation in EU law comes in the area of customs tariffs. Here, we have various headings and subheadings of the Combined Nomenclature of the Common Customs Tariff that have been interpreted and clarified not only by the Court of Justice but also by the Commission in its Explanatory notes to the EU Combined Nomenclature, which are regarded as “an important aid to interpretation”. National courts often inquire about the exact tariff classification of certain products. For example, whether the term “seagoing vessels” covers vessels which, given their construction and properties, cannot sail beyond 21 nautical miles off the coast during bad weather; whether “Bob Martin Clear 50 mg – spot-on solution for cats”, a product for cats with specific chemical composition, packaging, mode of application, and effects should be classified as “a medicinal product” or “an insecticide”; or whether the expression “copper bars, rods and profiles” includes “copper or copper alloy ingots in a rectangular shape, the thickness of which exceeds one-tenth of the width and which are hot-rolled, but which have irregular pores, holes and cracks in their cross-section”.
For all these questions of interpretation of EU law, there arguably exists abundant case law of the Court of Justice. What will be left for the General Court, therefore, is to follow the abstract criteria specified by the Court therein and apply them to a new set of factual circumstances. But to have something to be guided by, questions of abstract interpretation which have not been addressed by the Court of Justice, should still be dealt with by that court.
Such a division of interpretive tasks between the General Court and the Court of Justice sits best with the structure of the preliminary ruling procedure. As is well known, in the division of work envisaged by art. 267 TFEU, the Court of Justice interprets EU law, and national courts apply EU law thus interpreted to specific facts. And the purpose of the procedure is to ensure uniformity of EU law in all Member States. In many situations, questions of factual interpretation coincide with what is usually understood as the application of EU law, which is the domain of national courts. In that respect, I agree with voices critical of the Court’s “factual jurisprudence”, which intrudes on this domain reserved for its interlocutors. As remarked by Advocate General Michal Bobek, “the primary role of the Court ought to be the articulation or the refinement of normative, legal premissa maior stemming from EU law, to be applied by national courts. The subsumption of the facts of the individual case, the premissa minor, and the conclusion as to the application of EU law in that particular case, is the task of national courts”.
In such a setting, uniformity that is the object of the preliminary ruling procedure concerns legal rules, which are products of the Court’s interpretation of provisions of EU law, and not outcomes of individual cases, which result from the application of those rules by national courts.
But national courts cannot be prevented from referring questions of factual interpretation for a preliminary ruling. Those may be genuinely difficult questions with which they need assistance. And it is clear that they have the discretion to decide if a(ny) question of EU law exists that needs to be clarified by the Court of Justice so they can deliver a judgment in the main proceedings. In that case, for the areas selected to come under the jurisdiction of the General Court, in the future those questions will be answered by the General Court only. As a rule, enough case law will exist to guide the General Court when providing answers to questions of factual interpretation. However, questions of abstract interpretation should be left to the Court of Justice. Uniformity of EU law at that level is the leitmotif of the preliminary ruling procedure. And the Court’s answers to those questions provide guidance to the General Court when answering questions of factual interpretation of EU law. So, the Court of Justice should remain responsible for those questions of abstract interpretation. This will be especially important in cases where the referring court questions previous abstract interpretations given by the Court of Justice, which are meant to guide the General Court. For instance, if the referring court has good reason to think it is appropriate to adjust or overrule the current interpretation of the concepts of “extraordinary circumstances” from the Passengers’ Rights Regulation or “school or university education” from the VAT Directive, the General Court should leave that decision to the Court of Justice.
VI. Which way, national courts?
If the General Court indeed replied only to questions of factual interpretation and if it left questions of abstract interpretation to the Court of Justice, as suggested here, this would solve important concerns raised by the proposal for reform of the preliminary ruling procedure regarding the role of national courts.
One of the important questions raised by the reform proposal is what will national courts do? Will they see the General Court as equally legitimate and competent an interlocutor as the Court of Justice and keep eagerly cooperating with it? Or will they try to tamper with their references by inserting arguments and information to mislead the Court of Justice into thinking that the “exclusivity” criterion is not met, so the reference should not be forwarded to the General Court?
With regard to the attitude of national courts towards the new form of the preliminary ruling procedure, the following should be mentioned. First, there is the principle of sincere cooperation from art. 4(3) TEU. Ideally, it should guide not only national courts in their interaction with the General Court, but should also guide the two EU courts in determining their respective jurisdictions and allocation of preliminary references between themselves. Their engagement should be respectful, and they should be doing everything to ensure the smooth functioning of the preliminary ruling procedure, and, by extension, the effectiveness of EU law and the attainment of the Union’s goals. All courts in the EU have a stake in this, and there is no way around cooperation, no matter how difficult it may be. The Court of Justice often emphasises the cooperative nature of the procedure, and that its ultimate purpose is the “administration of justice” in the Member States on the basis of EU law. So, the entire system depends on the bona fide behaviour of all the judicial actors involved. Second, there are pragmatic considerations. Specialisation of the General Court in specific areas might make it a very desirable interlocutor in the long run, especially if it manages to reply to preliminary references more quickly than is currently the case. The waiting time was always one of the main reasons why some national courts chose not to refer questions for a preliminary ruling. So, expertise plus response time and the efficiency of the General Court might lead national courts to prefer having it answer their questions, especially since the areas of EU law in which the General Court will have jurisdiction often involve complex, technical issues. In that respect, national courts with specialised jurisdiction, e.g. financial, commercial, and administrative courts, might prove particularly ready for collaboration with the General Court. Third, there are strategic reasons. So far, different motives and interests have been reported as driving national courts to engage in the preliminary ruling procedure. Some are very strategic about this. They may wish to “outsource” decisions on contentious points of EU law. For instance, when they have to go against the domestic executive or legislator, or circumvent their higher courts, national courts will still have an interest in reaching out to the General Court and using its decisions as a “sword” or a “shield”. Any kind of external validation is arguably better than having to face a potential backlash barehanded. For others, the fact that EU law formally provides an option or obligation to refer is enough. When they need assistance with the interpretation of EU law, they will earnestly request it. For national courts who firmly hold this legal formalist view, little with change after the General Court acquires jurisdiction in specific areas. Of course, there will be other factors that will influence the engagement of national courts with the General Court, including the general attitude of judges towards EU law and integration, their knowledge of and previous experience with EU matters, the persuasiveness of parties in domestic proceedings when they argue that a preliminary reference should (not) be submitted, etc.
Be that as it may, the proposal made here on how to distribute questions between the General Court and the Court of Justice in the selected areas might ease some of the existing concerns. As already mentioned, questions of factual interpretation of EU law can be considered as something for national courts to handle themselves. We have already heard several calls from the Luxembourg benches for greater restraint from national courts when referring questions that are about singular and non-recurring situations of narrow scope, which have no general or structural impact on EU law. In those situations, national courts arguably have enough guidance about the meaning of EU law and should be confident enough to determine whether certain facts are “subsumed” within that meaning, without asking for assistance. But should they decide to make a reference anyway, they will immediately know that it will be answered by the General Court. So, their dilemma will be either to take responsibility for solving the question of factual interpretation of EU law on their own, or to willingly let the General Court do the job for them.
If national courts generally took more EU law matters into their own hands, the two EU courts would be less burdened. This is so because the root of the malaise lies on the “input” side – the volume of references sent by national courts. For better or worse, the Court of Justice this time did not choose to approach the reform of the preliminary ruling procedure by turning to its “input” side, although certain opportunities have recently presented themselves. Rather, the Court chose to change itself, not the world.
It remains to be seen how durable this reform approach will be. A constant increase in the number of references is recognised by the Court of Justice in several places in the reform proposal. One can expect the upward trajectory to continue, especially since the increase in legislative production coupled with more knowledge and experience with EU law on the part of private parties, lawyers, and national courts can only mean more references in the future. As former president of the Court of Justice, Ole Due, warned when the idea was first floated: “[t]he proposed transfer to the [then] Court of First Instance of preliminary ruling cases within certain special areas may slow, but certainly will not stop the steady increase in the number of preliminary references made to the Court of Justice”. So, the new cases are likely to keep coming, increasing the workload and prolonging judicial proceedings. Both the Court of Justice and the General Court will perhaps be in need of reform again in the not-so-distant future. By not addressing the “input” side of the preliminary ruling procedure, the current reform proposal seems to be merely kicking the can down the road. But in a decade or so, it might be time to go down the rabbit hole and look for solutions among the “ordinary courts” of “general jurisdiction” in EU law.
This will take an enormous amount of trust and confidence in national courts and in their maturity and capability to handle points and questions of interpretation of EU law on their own. Interesting proposals based on a similar animus have been on the table for a while now, such as, for example, the “green light” system under which the referring courts would send reasoned proposals of the answers to preliminary questions and the Court of Justice would agree or disagree with them by delivering rulings in which it refers specifically to the reasons given by the national courts in the manner of an appellate court.
In the meantime, some national courts are trying to be heard by using the option provided to them in the Recommendations of the Court of Justice to national courts in relation to the preliminary ruling proceeding, wherein they are encouraged to express their views on the interpretation of the questions of EU law they refer. Some authors have found that not many national courts, especially high courts, have shown much eagerness to take the initiative in this way. However, recent research indicates that the number of national courts that are proposing more or less elaborated answers to the preliminary questions is not negligible. The same holds for the number of national courts that are proposing the right answers, i.e. answers that are eventually endorsed by the Court of Justice. National courts that refer more and become increasingly motivated to propose answers to preliminary questions will over time accumulate greater knowledge and experience, which will likely make them better with the interpretation of EU law and more confident that they can handle this on their own. Perhaps in this way, through organic growth and development instead of top-down reforms and ordering, the “input” side of the preliminary ruling procedure will eventually take care of itself.
The reform of the preliminary ruling procedure will likely redefine the roles of the Court of Justice and the General Court 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. The Court of Justice would thus come closer to a proper Kelsenian constitutional court, which is tasked with the authoritative determination of the meaning of EU law, in particular concerning the questions of abstract interpretation, and concerned primarily with the uniformity and coherence of that law at the general level. In the selected areas of law, at least as a first step, the Court of Justice should lose the final say over questions of factual interpretation, such as in which tariff box to put frozen camel meat, pyjamas, and nightdresses. These questions, again in the selected areas, should become the province of the General Court, which should be responsible for correcting individual outcomes and assisting national courts with questions of factual interpretation, which are a prerequisite for the application of EU law.
Before any of this can happen, let us first see what the two courts and the three legislators will do with the proposal for reform of the preliminary ruling procedure as it currently stands. Since they are in largely unchartered territory, it seems safe to assume that anything is possible. EU legal scholars will undoubtedly do some charting and navigating of this territory to help the courts and legislators get there. This Insight was a modest contribution to the same end. Others will hopefully join soon, and not only in this forum.
* Lecturer and Postdoctoral Researcher at the University of Zagreb, firstname.lastname@example.org.
 Request submitted by the Court of Justice pursuant to the second paragraph of Article 281 of the Treaty on the Functioning of the European Union, with a view to amending Protocol No 3 on the Statute of the Court of Justice of the European Union (‘the 2022 Reform Proposal’). The other part concerns the mechanism by which the Court of Justice decides whether an appeal against decisions of the General Court can proceed.
 Opinion 2/13 Accession of the European Union to the ECHR ECLI:EU:C:2014:2454 para. 176.
 Case C‑284/16 Achmea ECLI:EU:C:2018:158 para. 37.
 P Craig and G de Búrca, EU Law. Text, Cases, and Materials (Oxford University Press 2020) 496.
 It reads: “The General Court shall have jurisdiction to hear and determine questions referred for a preliminary ruling under Article 267, in specific areas laid down by the Statute”.
 In my understanding, it is uncertain whether the courts from “new” Member States brought an increase only in absolute numbers (i.e. the total number of references) or in relative numbers as well (i.e. the number of references per court). To that end, any difference in the referral rates between “old” and “new” Member States by now seems minimal, if it exists at all. M Broberg and N Fenger, Preliminary References to the European Court of Justice (Oxford University Press 2021) 29 ff.
 But see H Rasmussen, ‘Remedying the Crumbling EC Judicial System’ (2000) CMLRev 1071, 1098–1103; and T Tridimas, ‘Knocking on Heaven’s Door: Fragmentation, Efficiency and Defiance in the Preliminary Reference Procedure’ (2003) CMLRev 9, 20–21.
 For a discussion of all these issues, see A Dashwood and A Johnston (eds), The Future of the Judicial System of the European Union (Hart Publishing 2001), in particular contributions by P Lindh, ‘The Court of First Instance: Meeting the Challenge’ in A Dashwood and A Johnston (eds), The Future of the Judicial System of the European Union cit. 13; A Arnull, ‘Judicial Architecture or Judicial Folly? The Challenge Facing the EU’ in A Dashwood and A Johnston (eds), The Future of the Judicial System of the European Union cit. 41; A Dashwood and A Johnston, ‘Synthesis of the Debate’ in A Dashwood and A Johnston (eds), The Future of the Judicial System of the European Union cit. 55; O Due, ‘The Working Party Report’ in A Dashwood and A Johnston (eds), The Future of the Judicial System of the European Union cit. 87; A Dashwood and A Johnston, ‘The Outcome at Nice: An Overview’ in A Dashwood and A Johnston (eds), The Future of the Judicial System of the European Union cit. 219.
 S Iglesias Sánchez, ‘Preliminary Rulings before the General Court: Crossing the Last Frontier of the Reform of the EU Judicial System?’ (17 December 2022) EU Law Live Weekend Edition No 125 www.eulawlive.com.
 Regulation (EU, Euratom) 2015/2422 of the European Parliament and of the Council of 16 December 2015 amending Protocol No 3 on the Statute of the Court of Justice of the European Union. For a discussion, see D Sarmiento, ‘The Reform of the General Court: An Exercise in Minimalist (but Radical) Institutional Reform’ (2017) CYELS 236, 247 ff.
 Report submitted pursuant to Article 3(2) of Regulation (EU, Euratom) 2015/2422 amending Protocol No 3 on the Statute of the Court of Justice of the European Union (‘the 2018 Report’).
 Ibid. 3–4.
 Ibid. 7.
 The 2022 Reform Proposal cit. 3.
 The 2022 Reform Proposal cit. 5.
 Ibid. 5.
 Ibid. 6.
 Ibid. 7.
 Rules of Procedure of the General Court, art. 25(1): “The General Court may make one or more Chambers responsible for hearing and determining cases in specific matters”.
 The 2022 Reform Proposal cit. 7.
 Art. 20(5) of the Statute of the Court of Justice of the European Union: “Where it considers that the case raises no new point of law, the Court may decide, after hearing the Advocate General, that the case shall be determined without a submission from the Advocate General”.
 The 2022 Reform Proposal cit. 7. This new chamber formation is meant to avoid preliminary references being heard by the Grand Chamber of the General Court. The rationale of the Court of Justice was twofold. First, cases decided by the Grand Chamber of the General Court typically involve important questions of principle and, in the view of the Court of Justice, should be sent back to it as envisaged by art. 256(3) TFEU (discussed below). In this respect, art. 28 of the Rules of Procedure of the General Court (“Referral to a Chamber sitting with a different number of Judges”) provides that “[w]henever the legal difficulty or the importance of the case or special circumstances so justify, a case may be referred to the Grand Chamber […]” (para. 1). The proposal for a referral can come from “[t]he Chamber seised of the case or the President of the General Court […] at any stage in the proceedings, either of its or his own motion or at the request of a main party […]” (para. 2). The proposal is made to the plenum of the General Court, which then decides whether to refer the case to the Grand Chamber (para. 3). Second, given the number of judges sitting in the Grand Chamber of the General Court, it could turn out that judges that are not from the specialised chambers end up deciding the preliminary references. In the view of the Court of Justice, such a possibility would undermine the logic of having specialised chambers of the General Court exclusively handling preliminary references. For a critical appraisal of the Court’s reasons, see S Iglesias Sánchez, ‘Preliminary Rulings before the General Court: Crossing the Last Frontier of the Reform of the EU Judicial System?’ cit. 11–12.
 The 2022 Reform Proposal cit. 6.
 Arts 62, 62a, and 62b of the Statute of the Court of Justice of the EU; for a discussion, see S Iglesias Sánchez, ‘Preliminary Rulings before the General Court: Crossing the Last Frontier of the Reform of the EU Judicial System?’ cit. 13–14.
 The 2022 Reform Proposal cit. 4.
 Ibid. 4.
 Ibid. 5.
 The 2022 Reform Proposal cit. Annexes 2 and 3.
 The 2022 Reform Proposal cit. 5.
 Council Directive 2006/112/EC of 28 November 2006 on the common system of value added tax.
 Council Regulation (EEC) 2658/87 of 23 July 1987 on the tariff and statistical nomenclature and on the Common Customs Tariff, last amended by Commission Implementing Regulation (EU) 2022/1638 of 20 September 2022 amending Annex I as regards subheading 950510 (Articles for Christmas festivities).
 Regulation (EC) 261/2004 of the European Parliament and of the Council of 11 February 2004 establishing common rules on compensation and assistance to passengers in the event of denied boarding and of cancellation or long delay of flights.
 Rules of Procedure of the Court of Justice, art. 60(1) (“Assignment of cases to formations of the Court”): “The Court shall assign to the Chambers of five and of three Judges any case brought before it in so far as the difficulty or importance of the case or particular circumstances are not such as to require that it should be assigned to the Grand Chamber […]”. However, it is not always certain what the reasons are that merit the intervention of the Grand Chamber; see M Bobek, ‘What Are Grand Chambers For?’ (2021) CYELS 1.
 See above art. 20(5) of the Statute of the Court of Justice of the EU.
 The 2022 Reform Proposal cit. 5.
 See ‘Report by the Working Party on the Future of the European Communities’ Court System from 18–19 January 2000 (“The Wise Persons’ Report” or “The Due Report”)’ in A Dashwood and A Johnston (eds), The Future of the Judicial System of the European Union cit. 145.
 S Iglesias Sánchez, ‘Preliminary Rulings before the General Court: Crossing the Last Frontier of the Reform of the EU Judicial System?’ cit. 16.
 See the 2022 Reform Proposal cit. Annex 3, Tables 1 and 3.
 Consider, for instance, the number of preliminary references generated during 2021 in the following areas: agriculture (15), freedom, security and justice (102), competition (15), consumer protection (50), intellectual and industrial property (23), public procurement (18), social policy (46), and transport (50); see further Court of Justice of the European Union, Annual Report 2021 228–258.
 Case 26/62 Van Gend en Loos v Administratie der Belastingen ECLI:EU:C:1963:1.
 Case C-617/10 Åkerberg Fransson ECLI:EU:C:2013:105.
 That being said, it can be acknowledged that for some these two might not be the most representative for the present purpose. Still, the Court of Justice in its case law database (InfoCuria at curia.europa.eu) determines the “subject-matter” of Van Gend en Loos as “free movement of goods” (primary area); “customs union” (secondary area); besides, the Court itself recognised Åkerberg Fransson as an example of a case which, appearing at first to deal with a particular area of law (VAT), concerned a larger structural issue of EU law (application of the Charter); see the 2018 Report cit. 5 fn. 12. Therefore, my hope is that, albeit imperfect, these two cases could be taken as an illustration of this point.
 I should point out that, in reality, this might happen only in a smaller number of cases. It is true that art. 94 of the Rules of Procedure of the Court of Justice (“Content of the request for a preliminary ruling”) requires national courts to summarise the subject matter of the dispute and relevant facts, identify applicable national law, and explain the reasons which raise doubt about the interpretation or validity of EU law and/or its bearing on national law. If national courts fail to do so properly, the Court may reject their references as inadmissible. However, in many preliminary references, a number of actors appear only when the matter comes before the Court of Justice; usually, the Commission and national governments are those that intervene, yet they do not take part in the domestic proceedings. In comparison to national courts, the Court of Justice will also benefit from the expertise of its research and documentation service. Hence, different points of view on the issue in question will regularly be brought up only in the proceedings before the Court. With all these additional inputs, the subject matter and importance of the reference may turn out to be quite different from what was initially considered by the national court. Moreover, it is not uncommon for the Court, for the same reasons, to rephrase the referred questions and address points that were not necessarily raised by the referring court.
 S Iglesias Sánchez, ‘Preliminary Rulings before the General Court: Crossing the Last Frontier of the Reform of the EU Judicial System?’ cit. 9–10.
 R Guastini, ‘Legal Interpretation. The Realistic View’ in M Sellers and S Kirste (eds), Encyclopedia of the Philosophy of Law and Social Philosophy (Springer 2020), who uses the terms “text-oriented” or “in abstracto” interpretation and “fact-oriented” or “in concreto” interpretation.
 The example was popularised by a famous Hart–Fuller debate; see HLA Hart, ‘Positivism and the Separation of Law and Morals’ (1958) HarvLRev 593; and L Fuller, ‘Positivism and Fidelity to Law – A Reply to Professor Hart’ (1958) HarvLRev 630.
 The 2022 Reform Proposal cit. 11: “[T]he General Court is perfectly equipped to adjudicate on requests for a preliminary ruling in those areas, since their factual and technical context determines, to a large extent, the useful interpretation of the relevant provisions of Union law” (emphasis added).
 Ibid. 4–5.
 Similar examples could be found in other areas not included in the reform proposal, which could be added in some future revisions if the General Court and the Court of Justice settled on the practice of dealing with the questions of factual and abstract interpretation of EU law, respectively. The most obvious candidates would be the concept of “use of vehicles” under art. 3 of Directive 2009/103/EC of the European Parliament and of the Council of 16 September 2009 relating to insurance against civil liability in respect of the use of motor vehicles, and the enforcement of the obligation to insure against such liability, developed in the case law starting with case C‐162/13 Vnuk ECLI:EU:C:2014:2146, where subsequent references inquired whether specific uses of vehicles or manoeuvres are covered by the meaning of that concept, thus triggering the insurer’s liability; and the concept of “working time” under art. 2 of Directive 2003/88/EC of the European Parliament and of the Council of 4 November 2003 concerning certain aspects of the organisation of working time, developed in the case law starting with case C-151/02 Jaeger ECLI:EU:C:2003:437, where subsequent references asked whether different forms of “on call” and “stand by” duties of emergency personnel such as medical workers and firefighters, with their own specific factual circumstances, come within the meaning of that concept. Both are discussed in case C‑561/19 Consorzio Italian Management e Catania Multiservizi and Catania Multiservizi ECLI:EU:C:2021:291, opinion of AG Bobek, paras 139–149.
 See case C-549/07 Wallentin-Hermann ECLI:EU:C:2008:771 para. 23.
 Case C-74/19 Transportes Aéreos Portugueses ECLI:EU:C:2020:460.
 Case C-449/17 A & G Fahrschul-Akademie ECLI:EU:C:2019:202.
 Case C-373/19 Dubrovin & Tröger – Aquatics ECLI:EU:C:2021:873.
 Case C-48/19 X (VAT exemption for telephone consultations) ECLI:EU:C:2020:169.
 Case C-215/19 Veronsaajien oikeudenvalvontayksikkö (Computer centre services) ECLI:EU:C:2020:518.
 Case C-559/18 TDK-Lambda Germany ECLI:EU:C:2019:667 para. 26.
 Case C-192/19 Rensen Shipbuilding ECLI:EU:C:2020:194.
 Case C-941/19 Samohýl group ECLI:EU:C:2021:192.
 Case C-340/19 Hydro Energo ECLI:EU:C:2020:488.
 The 2022 Reform Proposal cit. 5.
 Case 107/76 Hoffmann-Laroche v Commission ECLI:EU:C:1977:89 para. 5; case 69/85 Wünsche v Germany ECLI:EU:C:1986:104 para. 12.
 See G Davies, ‘The Division of Powers Between the European Court of Justice and National Courts’ (2004) ConWEB – Webpapers on Constitutionalism and Governance beyond the State 1; and M Bobek, ‘National Courts and the Enforcement of EU Law – Institutional Report’ in M Botman and J Langer (eds), National Courts and the Enforcement of EU Law: The Pivotal Role of National Courts in the EU Legal Order. The XXIX FIDE Congress in The Hague, Congress Publications Vol 1 (Eleven International Publishing 2020) 61, especially 87–89.
 Case C-923/19 Van Ameyde España ECLI:EU:C:2021:125, opinion of AG Bobek, para. 56 (emphasis in original).
 Case C‑129/19 Presidenza del Consiglio dei Ministri ECLI:EU:C:2020:375, opinion of AG Bobek, para. 149: “[T]he uniformity sought is not and has never been at the level of the single outcome of each individual case, but at the level of the legal rules to be applied. This means that, logically, while there is a reasonable degree of uniformity of the legal rules, there may be diversity in terms of specific outcomes” (emphasis in original); and Van Ameyde España, opinion of AG Bobek, cit. para. 61: “[S]uch a degree of uniformity in terms of homogeneity in outcomes in individual cases is, I would dare to say, a myth. In fact, such uniformity is not even achieved in highly centralised national judicial systems which, in contrast to the role of this Court with regard to preliminary rulings, carry out extensive review of the decisions of the lower courts as to the correct application of the law in individual cases”.
 S Iglesias Sánchez, ‘Preliminary Rulings before the General Court: Crossing the Last Frontier of the Reform of the EU Judicial System?’ cit. 10–11.
 I would like to thank the anonymous reviewer for inviting me to address these points.
 Achmea cit. para. 37.
 Case 16/65 Schwarze ECLI:EU:C:1965:117 at 886; case 244/80 Foglia v Novello ECLI:EU:C:1981:302 paras 16–18.
 AM Slaughter, A Stone Sweet and JHH Weiler (eds), The European Court and National Courts: Doctrine and Jurisprudence (Hart Publishing 1998); and recently, J Krommendijk, National Courts and Preliminary References to the Court of Justice (Edward Elgar 2021).
 Case C-338/95 Wiener v Hauptzollamt Emmerich ECLI:EU:C:1997:352, opinion of AG Jacobs, paras 10–21, 54–65; Consorzio Italian Management e Catania Multiservizi and Catania Multiservizi, opinion of AG Bobek, cit. paras 147–149; Van Ameyde España, opinion of AG Bobek, cit. paras 49–62.
 S Iglesias Sánchez, ‘Preliminary Rulings before the General Court: Crossing the Last Frontier of the Reform of the EU Judicial System?’ cit. 5.
 Most importantly, the Court decided not to revisit its CILFIT jurisprudence (case 283/81 CILFIT v Ministero della Sanità ECLI:EU:C:1982:335), which concerns the art. 267(3) TFEU obligation of national courts of last instance to send references for a preliminary ruling; see case C‑561/19 Consorzio Italian Management e Catania Multiservizi and Catania Multiservizi ECLI:EU:C:2021:799.
 O Due, ‘The Working Party Report’ cit. 89.
 Case T-219/95 R Danielsson and Others v Commission ECLI:EU:T:1995:219, Order of the President of the Court of First Instance, para. 77.
 Case T-51/89 Tetra Pak v Commission ECLI:EU:T:1990:41 para. 42.
 ‘The Due Report’ cit. 168; Association of the Councils of State and Supreme Administrative Jurisdictions of the EU and Network of the Presidents of the Supreme Judicial Courts of the EU, ‘Report of the Working Group on the Preliminary Rulings Procedure’ (2007) www.aca-europe.eu 8–9; European Parliament Resolution of 9 July 2008 on the role of the national judge in the European judicial system, paras 30–31.
 Court of Justice of the EU, ‘Recommendations to national courts and tribunals in relation to the initiation of preliminary ruling proceedings’, para. 18: “The referring court or tribunal may also briefly state its view on the answer to be given to the questions referred for a preliminary ruling. That information may be useful to the Court […]”.
 R van Gestel and J de Poorter, ‘Supreme Administrative Courts’ Preliminary Questions to the CJEU: Start of a Dialogue or Talking to Deaf Ears?’ (2017) Cambridge International Law Journal 122.
 Depending on the area of law and research focus, those numbers range between one third and one half of all national courts that appeared in the preliminary ruling procedure in a given period; see K Leijon, ‘National Courts and Preliminary References: Supporting Legal Integration, Protecting National Autonomy or Balancing Conflicting Demands?’ (2021) West European Politics 510; A Wallerman, ‘Can Two Walk Together, Except They be Agreed? Preliminary References and (the Erosion of) National Procedural Autonomy’ (2019) ELR 159; D Petrić, Interpreting EU Law: Legal Reasoning of National Courts in the Preliminary Ruling Procedure (Doctoral Dissertation, University of Zagreb 2022).
 Several studies have shown that around half the time the referring court and the Court of Justice agree on the answer to the preliminary questions and about half the time they disagree; see A Wallerman Ghavanini, ‘Power Talk: Effects of Inter-Court Disagreement on Legal Reasoning in the Preliminary Reference Procedure’ (2020) European Papers www.europeanpapers.eu 887; D Petrić, Interpreting EU Law: Legal Reasoning of National Courts in the Preliminary Ruling Procedure cit.
 M van der Woude, ‘The Place of the General Court in the Institutional Framework of the Union’ cit. 26.
 Case C-559/10 Deli Ostrich ECLI:EU:C:2011:708.
 Case C-395/93 Neckermann Versand v Hauptzollamt Frankfurt am Main-Ost ECLI:EU:C:1994:318.
 Case C-338/95 Wiener v Hauptzollamt Emmerich ECLI:EU:C:1997:552; S Iglesias Sánchez, ‘Preliminary Rulings before the General Court: Crossing the Last Frontier of the Reform of the EU Judicial System?’ cit. 15–16.