Codifying the ‘most intergovernmental institution’

Compared to the other institutions, the European Council stands out for the few Treaty rules governing it and the degree of informality which characterises its work. The reasons for this lie in its history and purpose. Indeed, the institution was born as an informal meeting of leaders, similar to classic international summits, in response to the challenges faced by the Community in the 1960s: member states aimed to ‘repoliticise’ European matters by placing sensitive political and economic issues back in the hands of Heads of State or Government.Footnote ²¹ As proclaimed at the end of the first summit, the new body would have established an ‘overall approach to the internal problems’ of the Community based on the political impetus of member states’ leaders,Footnote ²² whose implementation would then be left to Community institutions.Footnote ²³

Crucially, despite its increasing relevance, the European Council’s evolution has not been driven by legal provisions: in fact, it first appeared in primary law only in 1992 with the Maastricht Treaty, which defined its tasks similarly to those provided now by Article 15 TEU.Footnote ²⁴ Over time, the European Council increased its influence through political reality – revising major Community policies, pushing Treaty reforms,Footnote ²⁵ or providing political momentum to respond to crisesFootnote ²⁶ – essentially developing autonomously from Community law.Footnote ²⁷ It was only the Lisbon Treaty that formally made it a Union institution, codifying a role already established in practice. Put different, the European Council’s development has been marked by the law’s catching up with arrangements that had been already established in practice: first came the initiatives of the Heads of States and Government to bring integration further through ‘old-style’ summits to counterbalance the weakly political, technocratic Commission. Subsequently, the Treaties formalised the its role in law: the European Council was integrated into the Union’s institutional structure, and its relationship with other institutions became more explicit.Footnote ²⁸ Still, the crucial character of the institution was maintained throughout this codification.

This is evident from the ‘Lisbon codification’: while the latter implied the recognition of the centrality of the role of the European Council in sensitive areas like the Common Foreign and Security Policy or economic policy, with provisions envisaging a specific role in decision-making in these areas, the drafters acknowledged the importance of informality to the work of the European Council and maintained a broad formulation to describe its general tasks.Footnote ²⁹ Hence, as seen, Article 15 TEU now provides that, without exercising legislative functions, the institution gives the Union ‘the necessary impetus for its development’ and ‘define[s] the general political directions and priorities thereof’.

Giving substance to these formulations is no easy task: admittedly, it is difficult to reconcile the European Council’s political aspirations (to give the Union the necessary impetus …) with a task that appears generic and potentially weak (… through general political directions).Footnote ³⁰ If one were to try to make sense of it, the European Council’s mandate may be understood as steering the Union’s policy priorities and fundamental political positions internally, determining ‘where the Union should go’, and defining where the Union stands in relation to the outside world externally. In this light, the European Council is expected to ‘give direction’ to the Union, set its priorities, provide the necessary political backing, and lay the groundwork for the instruments required to implement reforms. Yet its guidance should remain general in nature. Not only is this implied in the Treaty’s qualification of the European Council’s directions as ‘general’,Footnote ³¹ but it is also clear from a comparison of the various language versions. In French, the institution provides orientations,Footnote ³² and in Italian, orientamenti,Footnote ³³ terms that convey the idea of defining broad principles and general rules, from which subsequent legislative measures will derive their concrete implementation. A further confirmation of this understanding is supported, as will become clearer in the following section, by a systematic reading of the limitations placed on the European Council’s action.

This is not to say that the Lisbon codification necessarily weakened the European Council’s role for the Union. Its function of providing impetus continues to stem from the political authority its members wield in their respective national contexts and from the flexibility of its internal procedures. And indeed, still today, the European Council operates through the typical mechanisms of the international summit, far from the character of the other EU institutions. It has effectively remained ‘the most intergovernmental institution’.Footnote ³⁴

In effect, this expression highlights how its members, the member states’ Heads of State or Government, act as if they are in classical international organisations and leverage their political relevance to assert influence over other institutional actors. This makes Conclusions a form of soft law: though they are formally non-binding, they are worded similarly to hard law, aim to shape the behaviour of their addressees, and are highly influential in practice.Footnote ³⁵ The Council, composed of (lower-ranked) representatives of the same member states, remains ‘captured’ by the European Council; likewise, thanks to its political weight, other institutions often follow its indications, especially in moments of political or economic crisis.Footnote ³⁶

Another characteristic of European Council’s intergovernmental nature is the pattern of negotiations that take place there, and which are influenced by the member states’ domestic priorities. Moreover, the prominence of national interests is further strengthened by both the distribution of powers between its members, which allows bigger states to assert their dominance over the smaller ones,Footnote ³⁷ and by the Treaty rules still requiring unanimity for several decisions: consequently, single member states may leverage using their veto power to negotiate outcomes aligning with their preferences.Footnote ³⁸ Finally, member states remain naturally dominant in the European Council vis-à-vis the Commission President, despite the fact that (s)he is too part of the institution: this makes it difficult for her or him to advance positions against the states’ reluctance.Footnote ³⁹ The picture is completed by the omnipresent secrecy surrounding the summits, intended to ensure states’ negotiating freedom,Footnote ⁴⁰ and by an admittedly limited political accountability of the Council as a body.Footnote ⁴¹

These characteristics may be warranted when the Union needs ‘impetus for its development’: to an extent, it is suggested that this need for more intergovernmental cooperation is inevitable, given the balance of power between the Union and its member states, as portrayed in the Treaty framework.Footnote ⁴² However, the European Council’s intergovernmental nature contrasts with the rationale of other EU institutions and of other decision-making procedures characterised by openness and transparency.Footnote ⁴³ This is true, first, for the Parliament and the Commission, which embody a more supranational approach since they protect interests other than those of individual states. But this is so even for the Council, which operates with safeguards, such as openness and transparency in legislative deliberations, which are absent in European Council’s working methods.Footnote ⁴⁴ While this intergovernmental role is justified for its impetus function, it is crucial to prevent this rationale from contaminating decision-making in the other institutions. Fortunately, the Treaties provide safeguards in this regard, at least on paper.

The limits to European Council’s action: the Lisbon Treaty’s institutionalisation and the institutional balance

Outside the (few) cases in which its action is precisely defined in the Treaties, the European Council is subject to two general limits deriving from primary law and introduced when the Lisbon Treaty made it a fully-fledged institution of the Union. Thus, unlike the situation pre-Lisbon, following its ‘institutionalisation’ the European Council cannot be said to operate in an (almost) complete legal vacuum.

First, this is so because Article 15 TEU establishes a clear prohibition: namely, that the European Council ‘shall not exercise legislative functions’. Though it may seem straightforward, that provision could be read in two ways. A narrower interpretation would limit the prohibition to the formal adoption of legislative measures. In other words, that institution could adopt ‘legal’ acts but without legislative value,Footnote ⁴⁵ which the Treaties specifically reserve as the outcome of legislative procedures.Footnote ⁴⁶ Under this perspective, there would be no problem with the level of detail of Conclusions, as long as they do not have the same effects as legislation. Conversely, a more substantive reading extends the prohibition to all functions that other institutions perform in a legislative procedure.Footnote ⁴⁷ This means that, unless otherwise provided in the Treaties, the European Council should refrain from taking decisions on legislation, including the definition of the form and precise content of measures, which belongs to the Parliament and the Council. When assessing Article 15 TEU in context, the second interpretation is preferable. This is because, first, the provision itself uses a broad terminology (‘functions’) instead of more specific formulations, such as ‘legislative powers’Footnote ⁴⁸ or that the European Council ‘cannot adopt legislative acts’.Footnote ⁴⁹ Moreover, if the formalistic reading were adopted, then the prohibition would merely be a restatement of the other Treaty provisions that do not mention that institution in legislative procedures, rendering Article 15 TEU devoid of purpose.Footnote ⁵⁰ Therefore, at least post-Lisbon, to give concrete meaning to that provision requires acknowledging that the content of Conclusions cannot be unfettered but must be limited to ‘general’ directions and political indications.

Second, the European Council’s scope of action is further limited by the principle of institutional balance. Derived by the Court from Article 13(2) TEU, this principle aims to safeguard the arrangement, established by the Treaties, of the different interests protected by each institution.Footnote ⁵¹ According to Lenaerts and Verhoeven, it has three components: (i) that institutions are sufficiently independent; (ii) that they do not unconditionally delegate their powers to others; and (iii) that they do not encroach on others’ prerogatives while exercising their powers.Footnote ⁵² To be sure, the real scope and applicability of the principle is not entirely clear from case law, hence why more recent authors link the use of the principle of institutional balance to a more limited gap-filling function by informing the interpretation of vague Treaty provisions regulating the powers of institutions.Footnote ⁵³ Be that as it may,Footnote ⁵⁴ this is already useful when it comes to the European Council, as it can provide a more concrete meaning to the vague content of Article 15 TEU: it signals that its political directions must leave other institutions a substantial, and not purely symbolic, margin of action to implement them.

The picture is completed by the principle of mutual sincere cooperation under Article 13(2) TFEU. By establishing that institutions act, within the limits of their powers,Footnote ⁵⁵ ‘with due regard for the powers of the other[s]’,Footnote ⁵⁶ this duty seeks to establish a non-dominating dialogue between them, preventing any institution from encroaching on the others’ prerogatives.Footnote ⁵⁷

That said, the actual weight of the principle of institutional balance on the European Council has been sometimes contested. It has been argued, for instance, that when such institution adopts Conclusions and political declarations, it typically acts ‘outside the EU’s institutional balance’,Footnote ⁵⁸ as these actions are not covered by Treaty provisions regulating the adoption of Union measures. This may have been the case before the Treaty of Lisbon, as the European Council was not considered an institution. Its members could then effectively be seen as adopting decisions collectively as ‘masters of the Treaties’, thus falling outside the institutional balance.Footnote ⁵⁹ Yet, since the Lisbon Treaty institutionalised the European Council, this is no longer true, since member states act therein within an institutional framework that gives them limited (conferred) powers. In legal terms, they act as European Council, not as member states.Footnote ⁶⁰

Additionally, one could claim that the European Council is not bound by Article 13(2) TEU when it adopts Conclusions, since the latter are, strictly speaking, political and non-binding acts. However, EU institutions are always bound by EU primary law regardless of the nature of their acts, whether binding or non-binding, formal or informal. Indeed, as the Court has held, they cannot evade the limits of their attributed powers just by adopting non-binding measures;Footnote ⁶¹ the same applies to informal institutional practices, which cannot run contrary to the Treaties;Footnote ⁶² and finally, in Ledra, the Court established that Union institutions are bound by EU primary law even when acting outside the formal EU legal framework.Footnote ⁶³

Therefore, even when the European Council adopts declarations under Article 15 TEU, giving the Union general political directions, it remains bound by the principle of institutional balance. The Court finally confirmed this view in two judgments concerning European Council Conclusions, which will be fully analysed in the third part of this work. Suffice it to say here that, in those two cases, the Court ruled that allowing the European Council to interfere directly with the content of legislative measures, binding the EU legislature with its Conclusions, would run contrary to the principles of institutional balance and conferred powers.Footnote ⁶⁴ Of course, this does not mean that the European Council cannot adopt any position on an issue which is the subject of a legislative procedure,Footnote ⁶⁵ as that would render its role meaningless. Rather, it implies that, although giving the Union ‘the necessary impetus’ may involve significant intervention by the European Council, this should never cross the red line of exercising the prerogatives of other institutions.

How should this framework, then, be put into practice? Ideally, when EU leaders recognise that economic, political, or social circumstances require a certain (or stronger) action by the Union, the European Council can use its Conclusions to signal that there is support among the member states for such action. This initiates a process in which the Commission drafts a suitable legal proposal, whose actual content is then negotiated by other institutions: usually, in the ordinary legislative procedure, the Parliament and the Council.Footnote ⁶⁶ Nevertheless, it is submitted that the reality often differs from this ideal framework. As the next section will posit, the broad wording of the Treaties, combined with the necessities raised by the political situation, leaves the door open for the European Council’s true intergovernmental nature to influence the legislative process more profoundly than the Treaties would permit, thereby exercising powers not explicitly conferred upon it and infringing the principle of institutional balance. This happened notably in three situations, which will be analysed in turn.

The Multiannual Financial Framework

To begin with, perhaps the most striking example of how the European Council determines the content of measures that should be defined by other institutions is the Multiannual Financial Framework, i.e. the Regulation that sets expenditure ceilings for the Union’s annual budget over periods of at least five years. Although technically a budgetary instrument internal to the EU, the Framework has always been deeply influenced by member states, given their role as the Union’s primary financial contributors.Footnote ⁶⁷ This has been clear since, in the late 1980s, states within the European Council began negotiating informal ‘financial package deals’ to set spending ceilings and allocations, which would then be given legal effect through an interinstitutional agreement.Footnote ⁶⁸ With the Lisbon Treaty, the Multiannual Financial Framework also became codified in the TFEU. Interestingly, despite the clear role exerted by the European Council until that moment, the Treaty did not fully codify institutional practice, and that institution was left out from the formal procedure: Article 312 TFEU only provides that the Framework is adopted unanimously by the Council after receiving the consent of the Parliament.Footnote ⁶⁹ This notwithstanding, the European Council retained its dominant role even post-Lisbon.

In the negotiations concerning both post-Lisbon Frameworks – the 2014-2020 one, and the 2021-2027 one, it is possible to detect similar (problematic) patterns of action. As Crowe observed, the work of EU institutions seemed to be intended to prepare negotiating frameworks for the European Council to decide on.Footnote ⁷⁰ Indeed, on both occasions, drafts of the Multiannual Financial Framework would be sent by the Council to the European Council without figures, which were instead replaced by an ‘X’.Footnote ⁷¹ Only after the leaders’ summits would numbers appear,Footnote ⁷² in remarkably detailed and prescriptively-worded Conclusions.Footnote ⁷³ As a result, the Parliament, by political and economic pressure due to the expectation that the Framework would be adopted before the new budgetary period, essentially ended up ‘rubber-stamping’ a pre-cooked political agreement.Footnote ⁷⁴ It is true that for the Multiannual Financial Framework 2021-2027, the Parliament did manage to push some of its requests into the final Regulation.Footnote ⁷⁵ The fact remains, though, that the final Regulation was based on the figures once again defined in detail by the European Council. Other institutions were then left with no more than six months to complete the negotiations, under constant pressure to conclude the deal swiftlyFootnote ⁷⁶ – so much so that, eventually, legislative negotiations did not alter the monetary ceilings established in the Conclusions.Footnote ⁷⁷ Thus, in both instances, the European Council effectively managed to ‘attract’ the negotiations back to itself, maintaining the pre-Lisbon practice of a ‘two-phase decision-making process’:Footnote ⁷⁸ a first phase dominated by European Council bargaining; and a second formalising its political agreements. And recently, the midterm revision of the Framework to increase the Union’s support to Ukraine confirmed these negotiating patterns.Footnote ⁷⁹

In all these cases, the content of the Multiannual Financial Framework largely confirmed the Conclusions, with only minor adjustments at most. On top of other issues entailed by this practice,Footnote ⁸⁰ it is striking that this is already at odds with the institutional balance set by the Treaties. Although the latter do favour, in such long-term budgetary procedures, the member states’ interests over the Parliament – which has only power to consent – they provide that the Framework should result from a legislative process that does not involve the European Council, but only the Council and the Parliament. By involving itself in such detail in the adoption process, the European Council, far from giving ‘general’ political directions, appeared to act contrary to the institutional balance envisaged in the Treaties.

Perhaps even more blatantly, this practice contrasts with the prohibition on exercising legislative functions. In its case law related to the European Council, the Court has linked the legislative power, which the Treaties assign to the Council and the Parliament, to the power to decide the content of a measure.Footnote ⁸¹ Also, it has ruled that, in principle, institutions remain free to determine the content of legislative measures regardless of what Conclusions indicate.Footnote ⁸² However, if due to time constraints and the political sensitivity of the matter, the legislative institutions have no other choice but to confirm the substantive content reached by the European Council – because, say, the Parliament’s threats to deny consent are nullified by the inevitability and necessity of the Framework’s adoption – they are de facto deprived of their power to decide the content of the measure.Footnote ⁸³ In reality, then, it is the European Council that exercises legislative functions; functions that, as seen, do not belong to it.

Therefore, there seems to be a clear mismatch between the Treaties’ understanding of the adoption process of the Multiannual Financial Framework and the political reality. In this regard, it could be maintained that with such sensitive political choices, even detailed quantitative figures fall within the European Council’s function of providing ‘the necessary impetus’ for the Union’s development under Article 15 TEU.Footnote ⁸⁴ Yet this reading, albeit relevant, risks offering only a partial reading of the Treaties. As described previously, the principle of institutional balance is not ‘static’ but relational: the powers of one institution must be understood (also) in light of those conferred on the others participating in the same procedure. Consequently, even if detailed European Council guidance might be acceptable in situations where the final decision lies exclusively with the Council, perhaps even requiring unanimity, as in the case of the Own Resources Decisions establishing member states’ contributions to the EU budget,Footnote ⁸⁵ the same cannot be said for other situations like the Multiannual Financial Framework: where the Treaties require (at least) the Parliament’s consent, the European Council’s detailed involvement becomes problematic, as it prevents the legislature from exercising its role in a meaningful way. It is in such circumstances that the institutional balance risks being effectively altered.

In any event, some argue that this approach is unavoidable, as budget-setting requires balancing national concessions among 27 ‘naturally selfish’ member states, all of whom must agree due to unanimity rules.Footnote ⁸⁶ And given the control of governments on fiscal policies, only European Council proposals seem to stand.Footnote ⁸⁷ Yet, if that is the case, then it is puzzling why, although the political reality had been there all along, the drafters of the Treaty of Lisbon codified the institutional practice related to the Multiannual Financial Framework to such a limited extent. In light of this, this conflict between the (Treaty) law and practice remains alive and kicking.

A similar mismatch can also be observed in two other instances from 2020. First, in July 2020, European Council Conclusions made possible the birth of Next Generation EU, a set of tools to mitigate the economic impacts of the Covid-19 pandemic on member states. The second Conclusions, in December, addressed the adoption of the Rule of Law Conditionality Regulation, designed to protect the Union’s financial interests against rule of law deficiencies in member states.Footnote ⁸⁸ Crucially, both 2020 Conclusions were adopted in the context of necessity and urgency prompted by the pandemic, which represented a critical factor for the role assumed by the European Council.

Necessity requires action: the birth of Next Generation EU

When, early 2020, the Covid-19 pandemic broke out, causing major economic difficulties for member states as well as for the EU, the EU needed unprecedented effort to properly react, and it was felt that existing instruments to face economic crises were insufficient.Footnote ⁸⁹ Thus, the Commission proposed Next Generation EU, a package to disburse €750 billion to help member states’ economies recover. This package consisted of three different instruments, adopted through three different procedures ranging from the ordinary legislative procedure (for the Recovery and Resilience Facility, hereinafter the Facility) to the special procedure requiring unanimity in the Council (for the Own Resources Decisions).Footnote ⁹⁰ No mention is made in the Treaties of the European Council for the adoption of any of these acts. Therefore, based on the legal framework, one would expect that the details of Next Generation EU would be decided by the two legislative institutions, particularly that the most innovative mechanism, the Facility, would be agreed upon by the Parliament and the Council acting as co-legislators. Yet such expectations would be wrong.

Indeed, it was immediately clear that the member states would be at the centre of the negotiations and had chosen the European Council as the institutional setting in which to conduct them. Not by chance, the very first proposal to finance recovery measures with common debt was sent by nine Heads of State or Government to the European Council President,Footnote ⁹¹ who himself started the negotiations to come to a final agreement.Footnote ⁹² Then, in May, a Franco-German proposal defined the general character of a potential recovery instrument, including the possibility of financing it with Commission borrowing on the markets. Just a few days later, the Commission published its proposal for Next Generation EU, along the same lines of the Franco-German initiative.Footnote ⁹³

But the real significance of the European Council’s role became apparent in July, at the special summit convened to discuss Next Generation EU.Footnote ⁹⁴ In its Conclusions, a 67-page document, the European Council did something ‘very much like legislative drafting’.Footnote ⁹⁵ First, the Conclusions specified the precise amounts of each spending programme, divided into loans and grants.Footnote ⁹⁶ Furthermore, they extensively employed prescriptive terms like ‘shall’, which the Court has in the past deemed to be evidence of an institution’s behaviour ‘manifestly seeking to impose’ its will on other actors, especially when combined with the precision of the content of an act.Footnote ⁹⁷ Lastly, a whole section of the Conclusions was dedicated to what would become the Regulation establishing the Facility. Notably, the Conclusions detailed the precise management of Next Generation EU, devising an architecture with the Council ‘in the driving seat’ together with the Commission.Footnote ⁹⁸ On this point, even the ‘Community method’ components present in the Facility reflect the European Council’s role as the principal initiator and shaper of the instrument. Indeed, while it is true that the Commission, a typical supranational institution, has gained significant power in overseeing Next Generation EU national implementation, it is striking that the Parliament has been given no role at all. Relying on established practices related to the Union’s budget implementation, the two institutions began to work in close cooperation, to the detriment of the other supranational institution – the Parliament.Footnote ⁹⁹

It is true that the Conclusions were not the final word and the various elements of Next Generation EU were adopted following the formal procedures envisaged in the TFEU. Yet, just as with the Multiannual Financial Framework, the essential components of the package, such as the monetary ceiling and the approval procedures for national recovery plans, remained untouched.Footnote ¹⁰⁰ Following the Court’s definition of legislative functions,Footnote ¹⁰¹ then, it was the European Council that decided the content of the measure(s) at stake. Effectively, the institutional balance of decision-making process emerging from the Treaties appears shifted, in particular as regards the Regulation establishing the Facility: even though the latter was formally subject to the ordinary legislative procedure, the Parliament was sidelined to the advantage of the member states, as the substantive content remained mostly unchanged following the Conclusions.

To be sure, it is true that Next Generation EU was created in the context of an unprecedented emergency. Already in general constitutional theory it is believed that times of crisis require rapid and decisive responses that ‘cannot wait for the deliberate pace of ordinary constitutional rule’,Footnote ¹⁰² resulting in a (temporary) reinforcement of the executive power.Footnote ¹⁰³ Therefore, it is rational to believe that, given decision-making procedures that require compromises between the Parliament and the Council, Union institutions may prefer to indulge the member states’ leaders, when they are willing to negotiate, informally and swiftly, behind the closed doors of the European Council:Footnote ¹⁰⁴ as underlined by Van Middelaar, ‘necessity knows no law’.Footnote ¹⁰⁵

While an analysis of the adoption of Next Generation EU under the framework of emergency powers goes beyond the scope of this article, suffice it to note that uncertainties remain even considering the emergency context. Indeed, unlike several national constitutions, the EU lacks a comprehensive emergency framework, with Article 122 TFEU being the only proper ‘emergency power’ relevant to Next Generation EU.Footnote ¹⁰⁶ Notably, the European Council is absent from that article; instead, one could claim that the emergency nature of the provision is already reflected, from a procedural standpoint, by the increased role for the Council compared to the ordinary legislative procedure.Footnote ¹⁰⁷

To conclude, the European Council was undoubtedly essential in helping the Union’s resilience through the pandemic. However, this should not result in overlooking the institutional costs of its actions – costs that are also evident in the adoption of another measure during the same period: the Rule of Law Conditionality Regulation.

Vetoes require compromises: the Rule of Law Conditionality Regulation

After two years of deadlock due to the Council’s hesitant attitude,Footnote ¹⁰⁸ the ordinary legislative procedure for a regulation to protect the Union from rule of law violations by its member states regained momentum when the pandemic broke out and the European Council linked the negotiations on the rule of law conditionality instrument to those concerning the Multiannual Financial Framework 2021-2027 and Next Generation EU. Consequently, the July 2020 Conclusions contained a reference to the fact that a regime of conditionality ‘[would] be introduced’.Footnote ¹⁰⁹

Albeit in line with the task of providing the political legitimacy needed to overcome the stalemate in the Council, two elements stand out oddly in such Conclusions. First, it was stated that measures proposed by the Commission under the conditionality instrument would be adopted by the Council by qualified majority, contrary to the Commission’s original proposal, which envisaged reverse qualified majority.Footnote ¹¹⁰ As the Council was naturally bound politically to their chiefs’ will, and the Parliament considered it as if it were a fait accompli,Footnote ¹¹¹ this modification was finally reflected in the final text of the Regulation.Footnote ¹¹² De facto, then, the European Council modified the content of the legislative instrument, contrary to the prohibition on exercising legislative functions. Second, the July Conclusions mentioned that ‘the European Council [would] revert rapidly to the matter’.Footnote ¹¹³ Especially considering the reservations of some member states on a rule of law conditionality mechanism, this was a sign that the institution wanted to retain some leeway to ‘green-light’ the final Regulation.Footnote ¹¹⁴ Once again, it ‘attracted’ negotiations to its level.

When, that autumn, the Parliament and the Council finally managed to reach agreement on the text of the Regulation, Hungary and Poland were dissatisfied with the outcome of the institutional negotiations, inter alia, as it did not ‘conform to the agreement reached … at the July European Council’.Footnote ¹¹⁵ Legally speaking, this discontent would not have affected the adoption of the Regulation: the decision-making procedure required qualified majority, not unanimity, in the Council, and the Court had already made clear that due to the value of Conclusions as merely political, a piece of legislation could not be annulled simply because it diverged from them.Footnote ¹¹⁶ On the other hand, though, Hungary and Poland exploited the coincidence in time with the adoption of Next Generation EU and threatened to veto the Own Resources Decision – which required unanimity in the Council and ratification by all member states – if the text resulting from the legislative negotiations was approved without changes. Since such Decision was an essential component of Next Generation EU, failure to adopt it would have led to the collapse of the entire economic package.

At that point, Germany, which held the Council Presidency at the time, came up with a compromise to be discussed at the upcoming European Council meeting in December. The resulting Conclusions, ‘with a view to finding a mutually satisfactory solution’ on the conditionality mechanism,Footnote ¹¹⁷ included some rather controversial elements.

On the one hand, they emphatically reiterated elements already present in the (draft) Regulation to ‘reassure’ member states on the correct implementation of the mechanism through a sort of ‘authoritative interpretation’ of the text of the Regulation.Footnote ¹¹⁸ While this is clearly not the its role under the Treaties – in fact, this is the sole prerogative of the Court – these declarations did not have substantive implications for the Regulation and for its subsequent enforcement.Footnote ¹¹⁹ At least limited to these aspects, then, the content of the Conclusions was indeed ‘superfluous’.Footnote ¹²⁰

More problematic was, instead, a second type of declaration concerning the application of the Regulation. To begin with, the European Council tried to make the ‘emergency brake’ mentioned in the Regulation’s recitals more compelling, by expecting the European Council to formulate a ‘common position’ if a member state considered the application of conditionality to be discriminatory.Footnote ¹²¹ Although this mechanism remained in the recitals and did not influence the application of the Regulation, especially because the Court held that a recital cannot be relied on as grounds for derogating from the substantive provisions of the Regulation,Footnote ¹²² an informal involvement of the European Council in an ongoing rule-of-law assessment might still be problematic as it could diminish the Commission’s independence by exercising undue pressure on it.Footnote ¹²³

Moreover, the December Conclusions specified that before opening any proceedings under the Regulation a ‘thorough dialogue’ had to be established by the Commission with the member state concerned.Footnote ¹²⁴ While this statement did not result in a modification of the Regulation, it did influence the Commission’s approach towards it, as evidenced by its Guidelines on the application of the Regulation, which substantially incorporate the requirement outlined in the Conclusions.Footnote ¹²⁵

More problematically, the European Council ‘agreed’ that the Commission ‘intended’ to adopt Guidelines on how it would apply the conditionality mechanism and to wait until the conclusion of any action for annulment brought against the Regulation before finalising those guidelines and, crucially, proposing any measures against member states.Footnote ¹²⁶ In doing so, the European Council introduced the need for guidelines that were not envisaged in the text of the Regulation itself and were not initially foreseen by the Commission; also, while formally only the Court can order an interim suspension during judicial review, the Conclusions implied a de facto suspension of the application of the mechanism until the finalisation of the Guidelines and after the Court’s ruling.Footnote ¹²⁷ Besides encroaching on the Court’s prerogatives, this, as some argue, amounts to conditioning the application of a Regulation and de facto amending a legislative act, contrary to the prohibition on exercising legislative functions.Footnote ¹²⁸

Indeed, the Commission proposed the first conditionality measures in September 2022, almost two years after the adoption of the Regulation. This happened, in line with the Conclusions, only after the Court ruled on its validity and the Commission finalised the foreseen Guidelines.Footnote ¹²⁹ What is more, notwithstanding the Commission’s declaration that ‘if and when concerns arise, discussions with member states will commence without undue delay’,Footnote ¹³⁰ the Commission refrained from even starting the preparatory dialogue envisaged in the Regulation with Hungary until after the Court’s judgment.

To be sure, it is true that, as argued by the Commission, no real suspension was at stake, since rule of law concerns arising from the first day of application of the Regulation (1 January 2021) would eventually have been addressed.Footnote ¹³¹ However, even assuming that there was no suspension on the application of the Regulation, as the Commission maintained, there was at least a suspension of its enforcement. Indeed, the Commission did wait more than a year after the adoption of the Regulation to enforce it: as reported by Alemanno, Commissioners themselves believed that the Regulation could not be used until the Court had judged on its legality.Footnote ¹³² If one considers that, when sanctions for rule of law violations are concerned, the time factor is crucial not only in terms of their scope, but also of its enforcement,Footnote ¹³³ the difference between the suspension of application and of its enforcement is actually minimal.

In addition, the Commission also maintained, on the concerns over the Conclusions themselves, that where the Conclusions mentioned its intentions, they reflected the actual will of the Commission President. This would be confirmed by the explicit reference to the Commission’s intention to commit to, and apply, the Conclusions on the conditionality mechanism.Footnote ¹³⁴ According to De Witte, this very statement made the Conclusions less constitutionally problematic. He stressed that their wording, along with the fact that the Commission President is a member of the European Council, supports the view that the Commission decided to draft the applicative guidelines ‘out of its free political will’, and not because it was forced by the European Council.Footnote ¹³⁵ This argument, however, seems to overlook the significance of the rules concerning the Commission’s nature. The latter is designed to be independent from any other institutional actor: not only is this a constitutional requirement set in the Treaties,Footnote ¹³⁶ but it is also a necessary requisite for the Commission to be effective vis-à-vis the other institutionsFootnote ¹³⁷ and to protect the Union’s general interest in the long run. If, instead, its will is formed in a context dominated by the member states – in the European Council, precisely – one cannot be sure that its ‘own political will’ is genuinely ‘free’. In the specific case, the idea of suspending the application (viz, the enforcement) of the Regulation was openly put forward by Germany, not by the Commission, as a last resort to avoid Hungary and Poland’s veto. As was evocatively noted, the statement on the Commission’s intention had ‘the ring of an offer that [it] could not refuse’.Footnote ¹³⁸ In light of this, one could wonder what remains of the fundamental component of the institutional balance whereby the relationship between the institutions of the Union, especially in the case of a completely ‘supranational’ measure like the Rule of Law Regulation, should be one of non-domination.

To conclude, the Conclusions on the conditionality mechanism remain problematic. They amounted, in different ways, to an undue exercise of legislative functions; they resulted in its de facto suspension for almost two years, despite the Court’s prerogatives to decide on that matter; and they reflected an imbalanced situation where the Commission, with all probability, felt compelled to accept the compromise made in the European Council. While it may be true that there was no other political option and ‘it is delusional to believe that the Commission would have immediately moved to sanction countries after the regulation’s entry into force’ in any case,Footnote ¹³⁹ this pattern of action conveys a misleading view of the EU constitutional system: one in which member states can block unrelated discussions and anything is allowed in order to reach a compromise. As Baquero Cruz underlines, ‘this disingenuous emphasis on results over method is in stark contradiction with the essence of integration, which was always based on a primacy of method over results’.Footnote ¹⁴⁰

Interim conclusion: institutional (im-)balance?

The previous cases were just examples – the most grave, perhaps – of a recurring behaviour by the European Council.Footnote ¹⁴¹ Worryingly, when this occurs, the interests represented in decision-making shift to the advantage of member states and to the detriment of the more supranational institutions: the Parliament becomes marginalised, and the Commission is rendered instrumental to the compromises that the member states want or need. Consequently, what happens in practice does not reflect the (primary) law anymore.

Trying to put together the elements found in the previous sections, it is possible to identify the scenario in which the European Council appears, at least prima facie, to challenge the principle of institutional balance. This happens when two conditions are met: (i) first, the European Council exercises specific competences that the Treaties reserve for other institutions, such as legislative functions (like when it determines the substantive normative content of a legislative measure, as in the Multiannual Financial Framework and Next Generation EU) or decisions concerning the implementation and enforcement of a Union measure (as happened with the Rule of Law Conditionality Regulation); (ii) second, it does so in such a way that effectively deprives the relevant institution of the possibility of exercising its prerogatives, by leveraging legal (e.g. unanimity rules) and extra-legal elements (e.g. political pressure and time factor). The combination of these factors allows the European Council to impermissibly ‘deprive the other institutions of a prerogative granted to them by the Treaties themselves’.Footnote ¹⁴²

Contrary to what is often claimed, that the ‘real power relations’ between institutions analysed above would ‘not contravene the EU legal framework if there is no violation of formal procedural [Treaty] rules’,Footnote ¹⁴³ the previous sections have instead argued that the opposite is the case. Though, admittedly, EU legislation is still adopted and applied, formally speaking, by the ‘right’ institution(s) and through the ‘right’ procedures, what happens in reality is often different: the European Council seems to violate institutional balance, as well as other principles stemming from primary law in a way that is difficult, to say the least, to reconcile with the Treaties.

At this point, the reader might wonder whether the Treaties, as the Union’s ‘constitutional charter’, could (or should) be interpreted with a degree of flexibility in light of the European Council’s crucial mandate. After all, providing the ‘necessary impetus’ to the Union’s development might indeed entail a precise definition of contentious aspects of legislation and policies in such sensitive areas.Footnote ¹⁴⁴ This touches upon a long-standing question in constitutional theory: the relationship between (constitutional) law, politics, and the actual practice of constitutional actors. Scholars have long analysed, for instance, institutional practices as means to supplement and develop the often-vague wording of constitutional provisions;Footnote ¹⁴⁵ and the practice analysed in this article could simply be viewed as one such manifestation. To better analyse the implications of this claim, it is worth recalling the prescriptive aspirations of constitutional charters.

Of course, Treaty rules, like constitutional norms, can be stretched to a certain extent to accommodate political reality: that is inherent to the very nature of constitutions, which can hardly disregard the reality of the political system they preside over – that is, of their ‘material constitution’.Footnote ¹⁴⁶ However, this flexibility has limits: the accommodation should not go as far as emptying the Treaties of their normative force – for example, when other institutions are deprived of a meaningful opportunity to exercise their mandate, or when the European Council assumes functions that the Treaties explicitly withhold from it. Otherwise, if the political constitution were to effectively prevail over the legal constitution,Footnote ¹⁴⁷ that would question the extent to which the system is ultimately rooted in the value of the rule of law. Admittedly, this view is in line with the logic of legal constitutionalism.Footnote ¹⁴⁸ Advocates of political constitutionalism, conversely, would contest its premise, since their approach rests on the primacy of the democratic process over written constitutional law.Footnote ¹⁴⁹ Now, fully addressing the issue would require a much broader analysis, which unfortunately goes beyond the limits of this article. That said, it could be questioned how far the premise of political constitutionalism can apply to the European Council, considering the extent to which its democratic legitimation as an institution is often questioned.Footnote ¹⁵⁰

As a result, one should be cautious in portraying the European Council’s institutional practice as a case of ends (or ‘mandate’) justifying means. Such an interpretation comes at a constitutional cost: the erosion of the independence and prerogatives of other institutions. Accepting this institutional practice as a form of accommodating Treaty rules to the constitutional reality of the EU would effectively risk rendering the prescriptive side of the Treaties meaningless.Footnote ¹⁵¹

In light of these concerns, can the Court offer adequate protection to the principle of institutional balance?

Political, not legal: protecting the institutional balance

The Court has not often had the opportunity to rule on the impact of European Council Conclusions on the Union’s institutional balance. However, in two significant cases, which have been mentioned throughout this article and will be fully analysed in a moment, the Court was called upon by some member states in actions for annulment, claiming that certain Union measures were illegal as they diverged from the content of previously adopted Conclusions on the same matter. As will be seen, the Court rejected such claims.

In Slovakia and Hungary v Council, the two states challenged a Council Decision concerning the relocation of asylum seekers across member states in response to the 2015 migrant crisis. During the negotiations, the European Council had agreed that any decision on relocations should be made by consensus – despite the legal basis requiring a qualified majority – and would concern a maximum of 40,000 applicants until the European Council discussed the issue again. However, after a first Decision that strictly followed the Conclusions, the Council adopted a second one by qualified majority, concerning the relocation of an additional 120,000 asylum seekers, without having received any indication from the European Council. Slovakia and Hungary claimed that this was an infringement of an essential procedural requirement, thus making the second Decision invalid.Footnote ¹⁵² Yet the Court did not agree. It straightforwardly held that the Conclusions in question only displayed effects of a political, rather than legal, nature and hence could not be a ground on which the contested decision could be annulled.Footnote ¹⁵³ Indeed, it followed from the principles of conferral and institutional balance that the European Council could not: (i) make the Commission’s legislative initiative conditional upon it having agreed on it; (ii) alter the voting rules established by the Treaties unless following the specific procedures to do so; and (iii) unilaterally determine the subject matter, objective and content of Union measures.Footnote ¹⁵⁴

Similarly, Poland v Parliament and Council involved a piece of legislation that had set the start date of a mechanism reducing emissions two years earlier than the relevant Conclusions had indicated. Therefore, according to Poland, they had ‘encroached upon the powers’ of the European Council (sic).Footnote ¹⁵⁵ Once again, the Court recalled the primary importance of securing the power of deciding the content of a measure to the Parliament and Council, as required by the principles of conferral and institutional balance.Footnote ¹⁵⁶ Accordingly, letting the European Council make orders to the other institutions concerning the details of a measure would result in an illegal interference in the legislative sphere. As in the previous case, the Court found that the political effects of Conclusions could not be grounds for claiming the invalidity of a Union act.Footnote ¹⁵⁷

Both cases concerned Conclusions indicating a precise political direction and acts of other institutions that, after negotiations, (partially) went in the opposite direction. In essence, the Court ruled in both cases that, regardless of what the Conclusions state, such statements cannot bind other institutions and instead are confined to a ‘mere’ political level. Put differently, they are ‘essentially policy documents’ without binding legal force.Footnote ¹⁵⁸ Therefore, the other institutions remain free to decide as they deem fit, whether in line with Conclusions or not. The Court appropriately linked its outcome with the principle of institutional balance, confirming the non-dominating relationship that EU institutions should maintain with one another. It matters little whether the topic is politically sensitive for member states,Footnote ¹⁵⁹ or whether the political weight of European Council members is significant: to ensure that other EU institutions can effectively exercise their prerogatives, without ‘rubber-stamping’ political agreements taken outside Treaty-based procedures,Footnote ¹⁶⁰ it is necessary to accord Conclusions no formal legal value.

At first sight, this case law successfully manages to protect the balance of powers set out in the Treaties when other institutions have enough time and (political) resources to negotiate a different deal than had been envisaged in the Conclusions. The same line of reasoning, however, could worryingly backfire when the situation is the opposite – meaning when, as seen above, Conclusions are so detailed, compelling, and pressing that institutions are left with no other choice but to fulfil them in full.

A double-edged sword: not protecting the institutional balance

Imagining a hypothetical action for annulment against Conclusions potentially exceeding the European Council’s powers,Footnote ¹⁶¹ the Court’s approach just discussed – of treating them as non-binding, or even lacking legal effects altogether – leads to undesirable consequences. These, reflected in both parts of the potential assessment, create a jurisdictional loophole and eventually contribute to undermining the very balance the Court claims to protect.

To begin with, a first issue arises in the admissibility stage as to how the Court has framed Conclusions in its case law. Under Article 263 TFEU, actions for annulment can be brought against acts ‘intended to produce legal effects’. To reject the argument that Conclusions bind the Council’s legislative action, however, the Court only framed the Conclusions in terms of their political authority.Footnote ¹⁶² While somewhat implicit, the Court seems to align with the widespread opinion that such Conclusions fall within the realm of soft law – they carry political weight but no legal effects whatsoeverFootnote ¹⁶³ – thus leaving them outside the admissibility scope of actions for annulment. This is all the more true considering that the Court, in its case law, has consistently added to this formulation that the effects must be not only legal, but also ‘binding,’ hence essentially linking the requirement of being intended to have legal effects to being legally enforceable. As a consequence, a measure could only be challenged under Article 263 TFEU, at least in theory, if it has a legal coercion mechanism.Footnote ¹⁶⁴ As should be clear by now, European Council Conclusions can influence the behaviour of other actors – sometimes even leaving them in practice with no other choice – but are surely not legally binding.Footnote ¹⁶⁵ Therefore, if the Court were to straightforwardly apply this interpretation, Conclusions could never be challenged and would be shielded from review by a formal veil of lack of binding effects, regardless of their potential factual effects.Footnote ¹⁶⁶

The issue is common to similar instances of soft law: in actions for annulment against soft law measures, the Court has shown to prefer a narrow reading of the requirements and maintained the equation to binding effects, despite the numerous pleas for a more substantive reading both by Advocates GeneralFootnote ¹⁶⁷ and literature.Footnote ¹⁶⁸ In particular, as suggested by Advocate General Bobek, an act should be deemed to display legal effects when it impacts ‘the legal situation of its addressees’ due to its content and overall purpose showing the intention to make others abide by it.Footnote ¹⁶⁹ Due to their formulation, the level of detail in their content, and the context in which they are adopted, Conclusions would arguably be included among the acts displaying legal effects based on this interpretation. Moreover, as discussed in the cases analysed in the previous section, the European Council (and the member states within it) can leverage the necessity associated with the adoption of a particular instrument or the unanimity rules in the Council to make Conclusions more ‘compelling’. This creates an (indirect) compliance mechanism vis-à-vis the other institutions:Footnote ¹⁷⁰ taking the rule of law conditionality mechanism as an example, the threat to veto the Own Resources Decision – had the Conclusions not been honoured – is, de facto, equivalent to a threat of sanction similar to that of formally binding measures. This made the Conclusions less and less ‘soft’ and more and more like ‘law’.

Moreover, one could add that the exercise of prerogatives in lieu of another institution is already impacting the latter’s legal situation. As Advocate General Sharpston argued in Swiss MoU, ‘the fact that one institution has taken [a decision] whereas the Treaties give powers to do so to another institution means that the act of taking the decision has legal effects (by usurping the powers of the second institution)’.Footnote ¹⁷¹ Even if, in the situations discussed above, the European Council does not formally adopt a decision (as was the case with the Commission Decision in Swiss MoU) the substance is not so different, considering how, in practice, the European Council determines the figures and specific functioning of Union instruments.

After all, an interpretation whereby European Council Conclusions may have legal effect for the purposes of Article 263 TFEU would not be entirely unorthodox. Not only has the Court repeatedly held that the assessment of the legal effects of a measure should be based on substance rather than on form,Footnote ¹⁷² it has also underlined that an act can display legal effects even if the institution that adopted it typically has no binding powers.Footnote ¹⁷³ In light of this, the Court has occasionally applied a more substantive reading of ‘legal effects’ and, for instance, has once considered an action for annulment against Council Conclusions admissible although they had no formally binding effects.Footnote ¹⁷⁴ Finally, in OIV, the Court held that the resolutions of an international organisation to which the EU was not even a party produced legal effects within the EU legal order simply because the EU legislature turned their content into EU legislation. While this evaluation was not made in the context of the admissibility of an action, hence the different (less stringent) scrutiny, it could represent an important precedent: it supports the claim that European Council Conclusions produce legal effects when the legislature translates them directly into legislation.Footnote ¹⁷⁵

It is true that there are possible solutions to avoid the admissibility issue. First, at times, the Court has evaded the issue of legal effects when analysing measures that clearly lacked them or were examples of soft law.Footnote ¹⁷⁶ In other cases, rather inconsistently, the Court has analysed the admissibility together with the merits because the question on legal effects ‘is inseparably associated with consideration of the content of [the act] and, consequently, with consideration of the substance of the actions’.Footnote ¹⁷⁷ This approach seems applicable to European Council Conclusions as well, as the issue of whether they force other institutions to act in a certain way concerns both the question of whether the European Council acted ultra vires and infringed the institutional balance, as well as the admissibility of the action.Footnote ¹⁷⁸ Last, an option could be to challenge the final instrument tainted by the Conclusions, like the Multiannual Financial Framework Regulation, precisely because of the undue interference from the European Council. The action would consider the Conclusions as (informal) preparatory acts that affect the validity of the final instrument,Footnote ¹⁷⁹ even if the two acts belonged to two different institutions.Footnote ¹⁸⁰

However, even assuming that the question of admissibility is overcome, the Court’s approach to Conclusions is also problematic in the substance. Indeed, the risk would be in extending to every scenario the position that the European Council cannot bind other institutions due to the purely political character of its Conclusions, as the General Court did in Tàpias v Council, thus underestimating the enormous weight that political influence can have in constraining the margin of action of other institutions. In Tàpias, a private party challenged a Regulation concerning institutional staff on the grounds that, inter alia, the Regulation was the result of an infringement of the legislative procedure. Specifically, the European Council had precisely defined some parts of the Regulation that the legislature ‘rubber-stamped’ into the final act, thus allegedly violating Article 15 TEU.Footnote ¹⁸¹ The General Court, however, disagreed, heavily relying on the two precedents previously analysed. It first noted that it is the task of the European Council to ‘carry out policy-making and coordinating functions’ and that the legislative procedure was (formally) followed in accordance with the Treaties.Footnote ¹⁸² Second, it held that Article 15 TEU ‘does not preclude the European Council from adopting a position on an issue which is the subject of a legislative procedure’, and lastly, it concluded by reiterating that, in any case, Conclusions only display a political effect that ‘cannot be a ground on which the Court may annul a contested decision’.Footnote ¹⁸³

This direct application of the precedents in the exact opposite situation – Slovakia concerned a piece of legislation allegedly in contrast with European Council Conclusions, while Tàpias involved one acritically implementing them – effectively crystallises in the legal realm, with the force a ruling can have, the situation shaped by the European Council potentially in violation of Article 15 TEU. In fact, one could wonder whether the European Council is just ‘adopting a position’ when it includes detailed and precise figures in its Conclusions, together with the indication that the final act ‘shall’ contain such figures, as happened in the analysed cases (and also in Tàpias).

Instead, the Court should have acknowledged that a legal act with binding force is not the only way an institution may violate the prerogatives of others. This is clear in many member states already, which provide not only for the reviewability of soft law,Footnote ¹⁸⁴ but also for specific actions (often before constitutional courts) to rule on horizontal conflicts of competence arising from an institution’s acts and behaviours.Footnote ¹⁸⁵ As a result, the European Council would not be free to adopt any position on any legislative act through its Conclusions regardless of their actual content, scope, and context, and the Conclusions’ coercive force would not be overlooked simply because they are deemed political in nature. This approach would not contradict the Court’s precedents: the effects of Conclusions, being prohibited by the Treaties from constraining other institutions, would still not be grounds for annulling a piece of legislation for diverging from such Conclusions. On the contrary, the suggested approach would be fully consistent with the Court’s position in the two judgments analysed above, as it would ensure that the European Council is prevented from using its de facto power to ‘interfere directly in the legislative sphere’ in a manner contrary to the principles of conferral and institutional balance.Footnote ¹⁸⁶

Nevertheless, one could note that stronger enforcement of Treaty rules by the Court vis-à-vis the European Council might risk hindering the effectiveness of EU action. After all, the Court has proven sympathetic to ensuring the smooth functioning of the EU even with creative institutional solutions.Footnote ¹⁸⁷ The proposed approach would, conversely, constrain the European Council’s actions and potentially lead to challenges whenever Conclusions may be deemed too detailed. Accordingly, the costs of remedying the European Council’s judicial vacuum would be eventually higher than the remedying problem itself. Should this issue, then, be resolved at all?

The need to fill the vacuum

The answer to that question must be positive. The EU constitutional order requires, in light of the principles underpinning the Treaties, a form of legal accountability for its institutions, especially in case of significant influence over Union measures. And the European Council is no exception.

Yet that institution appears to evade almost any form of accountability. As already noted, it is difficult to hold it politically responsible for its choices as an institution, due to the lack of control from both the European Parliament and national parliaments. Moreover, while there are alternatives to an action for annulment to hold the European Council legally accountable, these seem insufficient to mitigate the costs it imposes on the EU institutional structure. One option could be relying on the European Ombudsman, who has the mandate to investigate maladministration in the Union and may address some of the gaps in the European Council’s legal accountability.Footnote ¹⁸⁸ In fact, the Ombudsman has in the past dealt with several cases concerning the that institution’s transparency, particularly regarding access to documents and its President’s meetings with stakeholders.Footnote ¹⁸⁹ However, the transparency issue is only the tip of the iceberg and leaves out many of the concerns about an institution exercising prerogatives that do not belong to it – not to mention that the Ombudsman’s decisions are not binding and hence that compliance with its recommendations is not full.Footnote ¹⁹⁰ Similarly, other actions that could be used before the Court to remedy a situation caused by the European Council are all somewhat imperfect. In the case of the rule of law conditionality, for instance, the Parliament brought a case (later discontinued) against the Commission for failure to enforce the Regulation pending the Court’s assessment of the action brought by Hungary and Poland. Yet not only is the action for failure to act applicable to a very specific kind of interference, but it also has very high thresholds for success that undermine its utility.Footnote ¹⁹¹

Against this backdrop, one understands why the lack of judicial scrutiny is problematic, especially considering the importance of the values at stake. The EU legal order is inspired to the principles of liberal constitutionalism, among whose core foundations are the need for checks and balances and judicial review.Footnote ¹⁹² In a ‘Union based on the rule of law’, no institution can be above or outside the law and, ideally, no institution should evade, de facto, the safeguards that are in place to ensure that power is not exercised arbitrarily. At EU level, it is true that the Court values the effectiveness of EU action and is favourable to political solutions advancing European integration. As De Witte notes in Pringle, for instance, the Court confirmed the legality of inter se agreement between member states tasking European institutions with roles and functions.Footnote ¹⁹³ Another example comes from trilogues, the informal meetings between representatives of the Parliament, the Council, and the Commission to make legislative negotiations swifter and easier.Footnote ¹⁹⁴ In this respect, the General Court did not question their legitimacy and instead acknowledged the effectiveness and utility of that institutional practice.Footnote ¹⁹⁵

However, in each of these cases, the Courts posed safeguards to protect the EU’s institutional structure and values. In the former, it limited the scope of possible inter se agreements to ensure they did not contravene the principle of conferral. In the case of trilogues, since their closed-door nature challenged key principles of the legislative process such as openness and transparency, the General Court in De Capitani curtailed the secrecy of these meetings by securing access to their documents,Footnote ¹⁹⁶ thus ensuring the possibility of identifying who was responsible for specific legislative choices – an essential component of mature democracies.Footnote ¹⁹⁷ With the European Council, however, one gets the impression that it is only half subject to primary law: that institution is correctly unable to enforce its Conclusions judicially, but unfettered in exploiting circumstances to assert its dominance over certain decisions. What is more, safeguards like in the case of trilogues are structurally impossible when it comes to the European Council, as the secrecy of its meetings and the freedom of negotiation are the institution’s very raison d’être, as noted in the Introduction above. Therefore, it would be hard to imagine an equivalent ‘balance of interests’Footnote ¹⁹⁸ like in De Capitani: as long as the European Council keeps interfering with legislation to such a deep extent, these fundamental principles will be sacrificed.

In the long run, at least in the areas where the European Council is more active, the risk is to cause a ‘constitutional mutation’ of the institutional balance set by the Treaties, and thus of the interests reflected in decision-making as well as the channels of democratic accountability.Footnote ¹⁹⁹ Therefore, it becomes necessary to establish judicial review going beyond formal legal mechanisms and acknowledging substance over form.Footnote ²⁰⁰ Otherwise, limiting the protection of institutional balance solely to legal provisions would neglect the practical power dynamics between institutions, thus neutralising much of the prescriptive value of relevant primary law.

After all, as the Court itself has said, measures adopted by an institution could encroach on the powers of other institutions, or exceed the limits which have been set in the Treaties. If such measures were excluded from review, that ‘would lead to a result contrary both to the spirit of the Treaty and to its system’.Footnote ²⁰¹