Rule of Law Mechanisms: Remote Risk Prevention in a Disorganised Crisis Management

Abstract

Over the last fifteen years, the European Union has built up its regulatory response to the rule of law crisis. That framework seeks to address rule of law-related undesirable events depending on whether or not they have already occurred. As risk prevention has a limited role in the EU’s rule of law policy, undesirable events are mostly dealt with in the context of crisis management. Evaluation, reaction and conditionality mechanisms are all affected by issues relating to operability and legitimacy which reduce their overall performance. The resulting moderate efficiency of crisis management tools may still be improved by making use of the upgraded rule of law mechanisms in a systemic way.

I. Introduction

Reflections on the rule of law crisis in the European Union usually focus on Member States’ defiant attitude. The CJEU’s position in this regard is that Member States’ authority in respect of their national identity does not extend to the point of compromising the EU’s identity by an understanding of the rule of law that differs from law being indistinctly applicable to legal subjects as well as public authorities and serving both as a basis of action and as a means of action in a legal order.¹ A common and uniform understanding of the rule of law is not only relevant at the abstract level of values theoretically shared by all MS even before their admission to the EU.² When it comes to the day-to-day functioning of the EU legal order, ROL breaches in MS rattle fundamental principles such as mutual trust, mutual recognition, and effective judicial protection, the interconnectedness of which³ may ultimately lead to a crisis of implementation of EU law in all MS.

The “constitutional domino effect” that characterises the EU’s ROL crisis certainly entails a series of undesirable events for the EU legal order, which can be dealt with in two ways. On the one hand, if risk is to be understood as an undesirable event to be prevented ⁴, then an anticipative policy can help avoid rule of law-related events that have not happened yet. On the other hand, crisis being defined as a series of undesirable events to be mitigated, a reactive policy is called for in case rule of law-related events have already happened. In short, this translates to either risk prevention or crisis management.

For example, concerns had already been raised⁵ as to whether Hungary, i.e. a Member State subject to both Article 7 TEU proceedings⁶ and suspension of EU funds⁷ for systemic rule of law breaches,⁸ should be allowed to hold the Council presidency starting from July 2024. Then, the risk of seeing EU policies steered in unfavourable directions for EU institutions by representatives of a Union-sceptic MS’s government became more serious by the announcement of the European Council’s President to step down early,⁹ only to be reduced to its initial degree by Charles Michel’s withdrawal from the upcoming EU election less than three weeks later.¹⁰ In the absence of a legal instrument for the EU to deal with such an event,¹¹ both the withdrawal itself and the European Parliament’s call for an immediate reaction¹² to an emerging institutional risk, averted by chance, seemed more like a reaction to events rather than an anticipation of those events.

This paper argues that the observation made about the above example can be generalised to the EU’s regulatory response to the rule of law crisis in its present state. This is so because a great number of rule of law-related undesirable events have already happened which, therefore, are being targeted by rule of law mechanisms in a context that falls under crisis management. So, where does risk prevention fit in?

In response to this question, Section II goes through the existing rule of law mechanisms’ architecture and demonstrates that risk prevention appears at most as an incidental corollary to crisis management, but not as a goal on its own. Given this empirical assessment, Section III tests the efficiency of rule of law mechanisms for the purpose of crisis management. Finally, Section IV draws some conclusions regarding what can be improved in dealing with undesirable events in the context of the rule of law crisis.

II. Target regulation: Rule of law-related risk prevention as a negligible accessory to crisis management

Because of its evolutive nature, it would be obviously unrealistic to expect the normative framework of an “ever closer union among the peoples of Europe”¹³ to provide a prepared solution for every unprecedented or unforeseeable event. Yet, the EU’s regulatory response to ROL breaches suggests that the EU was caught off guard by situations arising in countries like Hungary or Poland. This is all the more surprising in light of the fact that EU values had not been strangers to contestation¹⁴ prior to the current ROL crisis even before “axiologically problematic” Member States’ first encounter with the entity that is known today as the European Union.

Turning to the 1990s, it is around the Soviet Union’s dissolution that the accession to the EU of the newly autonomous Eastern European states first becomes conceivable, and the time comes for the European Council to agree upon the so-called Copenhagen criteria, reiterated in today’s Articles 2 and 49 TEU and accompanied by sanctions to be activated according to the procedure of today’s Article 7 TEU.¹⁵ Accordingly, a state could only become a member of the EU if it meets both axiological (“stability of institutions guaranteeing democracy, the rule of law, human rights and respect for and protection of minorities”) and economic (“a functioning market economy as well as the capacity to cope with competitive pressure and market forces within the Union”) conditions.¹⁶

A shift of focus towards values within the integration project comes across in an amplified way during the decade that follows. A good example is the draft constitution destined to be the final product of the Convention on the Future of Europe, even though it ultimately failed because of two national referendums which took place in no other founding MS than France and the Netherlands¹⁷. It should be noted that the representatives of the candidate Member States did take part in this Convention during the summer of 2003.¹⁸ Therefore, the candidate Member States could even less be accused of not knowing what they had signed up for at the end of 2003 and what values they were supposed (obliged) to respect from 1 May 2004. This is usually one of the grounds for criticism addressed to MS in the specific context of ROL breaches, also known as the ratio legis of infringement proceedings. This is not disputed.

Underlying philosophies, however, do not necessarily translate into practical realities. As such, a major blind spot of the 2004 enlargement process lies in the trust that the signatory future Member States would actually respect the set of values they had decided to adhere to. The metaphor of marriage might prove illustrative here. Partners who marry of their own free will either believe in the success of their relationship so much that they dismiss the idea of a prenuptial agreement, or they prefer to have a safety net with the rainy days in mind to avoid losing even more time and money during a potential divorce. As both options have their merits, there is a priori nothing wrong with the EU’s decision at the time to act in the more wishful way towards its soon-to-be members and in accordance with the principle of mutual trust. Indeed, it is extremely risky to put everything on the mutual trust card without a safety net, while not doing so in the spirit of tolerance and diversity could give grounds for accusations of hypocrisy. Nonetheless, backing up the anticipated trust that the ex-Soviet states of Eastern Europe enjoyed by more substantial guarantees for the respect of the EU’s values – similarly to the way the case of Ukraine’s accession is being handled today¹⁹ and even if there were no (visible) grounds for mistrust in the early 2000s – might have helped avoid at least parts of the ROL crisis.

Easy to reason with hindsight, one might say. Yet, there are probably other, potentially delicate, economic and/or political reasons for the EU to have put its trust in the Eastern European countries concerned, the investigation of which goes beyond the scope of this paper. If the accession of both Hungary and Poland took place in good faith for that matter, then either EU leaders at the time had no capacity to measure the risks the 2004 enlargement would entail for EU values, including the rule of law, or those same leaders were aware of the possible consequences but chose to ignore them. In any case, and this is a crucial point, presumption that everything would go well does not equal prevention of what could not go well.

Everything did not go well. In a nutshell, not shying away from cynicism, some had assessed the cost-benefit ratio of belonging to the EU based on the generosity – or the naivety – thereof, and concluded that economic benefits, rights and living standards are still worth it even if the occasional obligations, fines and public humiliation are to be endured. To bring the sanctions up to a level comparable to the benefits, a series of heterogeneous instruments have been adopted over the last fifteen years.²⁰ Even though ROL breaches may be targeted based on three empirically discernable approaches, their common characteristic is that most breaches have already occurred by the time the mechanisms are implemented.

First, evaluation mechanisms are designed to assess various aspects of the rule of law and/or overall compliance with it within the Member States. The annual rule of law dialogue aims to make MS address ROL issues within the Council meeting in its different configurations.²¹ The EU Justice Scoreboard²² aims to give an annual overview of national justice systems based on several indicators.²³ The EU Rule of Law Framework aims to engage a “structured exchange” with the MS concerned through a three-stage process (assessment, recommendation, follow-up), at the end of which Article 7 TEU can be activated in case the follow-up is not satisfactory.²⁴ The Rule of Law Review Cycle aims to elaborate annual rule of law reports and country chapters on the state of the rule of law in the EU and in MS.²⁵ An advantage is that the findings of some of these assessments can be used for more general purposes, either as grounds for implementing other mechanisms or as a reference outside the EU’s normative ROL framework. However, risk prevention by evaluation mechanisms comes down to contingencies: these findings might pinpoint some factors that could help anticipate some undesirable events.

Second, primary EU law contains several provisions which can also be characterised as reaction mechanisms to ROL breaches. The only real sanction is contained in Article 7 TEU;²⁶ the other reaction mechanisms are in fact ordinary judicial remedies employed for rule of law crisis management purposes. In this context, Articles 258, 259 and 260 TFEU allow for rule of law protection measures to be imposed on the MS concerned at different stages of the infringement procedure.²⁷ As for Articles 263 and 267 TFEU, even though the purpose of an action for annulment or a preliminary reference is not to protect the ROL, they are not unrelated to it: action for annulment is an obvious manifestation of the principle of legality, while preliminary reference secures a uniform interpretation of EU law²⁸; together, they are supposed to represent the correct implementation of the ROL by the CJEU, mirroring the idea of effective judicial protection laid down in Article 47 of the EU Charter of Fundamental Rights and Article 19 TEU.²⁹ From a more practical point of view, when called upon to rule by an action for annulment or a reference for a preliminary ruling assorted with a ROL dimension, the Court is free to take the opportunity to “dynamically complete the treaties”³⁰ by clarifying, updating or expanding on the position of EU law in rule of law matters. Again, the limited scope of the Treaties regarding the rule of law, as well as the complex case law background that connects these treaty provisions to each other and to fundamental EU law principles, suggest that the occurrence of a crisis related to rule of law issues was not considered a real danger (or risk) when the Lisbon Treaty was adopted.

Third, conditionality mechanisms, which started to emerge in the context of a new EU strategy,³¹ subject the allocation of EU funding to the respect for certain rule of law elements.³² Although only subsidiary,³³ the general regime of conditionality allows for the adoption of measures capable of depriving MS concerned from large sums of payments³⁴ – that is, if financial dimensions can be found to ROL breaches in order to bring them within the scope of the regulation.³⁵ Beside this explicit conditionality, both the Recovery and Resilience Facility³⁶ and the European Semester³⁷ are characterised by implicit conditionality: though not aiming for it, their effect in practice can equal an allocation of funds based on the respect for ROL elements.³⁸ Whereas the RRF lists a series of mandatory rule of law mentions to be included in national recovery and resilience plans,³⁹ the country-specific recommendations in the context of the European Semester often contain ROL elements⁴⁰ that must be followed both on their own merits and because of the need for consistency between the Semester and the Facility.⁴¹ Then, the EU institutions, especially the Commission and the Council, respond to these programmes and plans in line with similar procedures but with varying margins of discretion.⁴² In terms of risk management, conditionality mechanisms stand perhaps closest to the idea of prevention: they represent a guarantee, or even an award, whereby payment is made if what EU law considers as an undesirable event does not materialise in the MS. Nonetheless, this remains a far-fetched interpretation and not an explicitly stated goal.

This quick overview of the EU’s regulatory response makes it difficult to argue that a ROL-related risk prevention policy would exist at the EU level. At least, such an intuition is not apparent from the current mechanisms’ design. Instead, the EU’s action in ROL matters tends almost exclusively to what can be summed up as crisis management. From the point of view of undesirable events, this general trend can still be regarded as acceptable as long as the EU’s crisis management eliminates or at least compensates for the deficit in terms of risk prevention.

III. Crisis management through moderately efficient rule of law mechanisms

For the efficiency of crisis management, the tools themselves must be efficient. As for ROL mechanisms, several issues related to their operability, i.e. the ease with which institutions can activate and implement a mechanism under procedural provisions, condition their performance, i.e. the achievement of the predefined goals summarised in the previous section. The following analysis proceeds in that same order and draws examples from the Hungarian case when necessary; the latter choice being justified by the idea that rule of law mechanisms’ efficiency is the most apparent against the most dissenting MS in the matter.

As a starting point, Table 1 indicates where to find the procedural provisions for each rule of law mechanism’s implementation.

Table 1. Procedural provisions for rule of law mechanisms’ implementation

The provisions listed in Table 1 identify twelve different forms of intervention that specify EU and MS institutions’ role in implementing ROL mechanisms. Further categorisation allows for these forms of intervention to be assigned to four types of competence. Accordingly, examination, evaluation and report relate to “assessment”; initiative and proposal relate to the “launch” of a procedural phase; from the narrowest to the largest margin of appreciation, approval, recommendation, determination, implementing decision, decision and judgment relate to “deliberation”; finally, information to the European Parliament and dialogue within the Council relate to “communication.” Table 2 visualises this categorisation.

Table 2. Types of competence and forms of intervention in rule of law mechanisms’ implementation

Table 2 ⁴³ displays a fixed distribution of the institutions’ competences depending on the approach in question. Unsurprisingly, evaluation mechanisms involve a predominant share of analysis, but an almost non-existent share of the other competences. This trend is reversed in reaction mechanisms: each deliberative competence sits in the hands of a single institution without any reference to assessment or communication competences that would help obtain the necessary input for decision-making. This imbalance seems to have been recognised and, to some extent, reduced in the design of the most recent ROL mechanisms, as conditionality mechanisms activate most or all types of competence. Now, aside from the fact that the three blocks are so conspicuously separate, which is not necessarily a good thing, some issues inferred from the contents of Table 2 should also be addressed.

One: assessments are either carried out by the Commission or based on the Commission’s findings. Where does the assessed data come from? The 2023 EU Justice Scoreboard’s introduction contains the most precise indications in this respect⁴⁴. Description of sources of data becomes vaguer in the annual rule of law reports the Commission elaborates in the context of the Rule of Law Review Cycle.⁴⁵ Concerning the two remaining evaluation mechanisms, both the EU Rule of Law Framework and the Council’s annual rule of law dialogue simply state that they are based on data available from other institutions.⁴⁶ Beyond the unlikelihood of MS’ cooperation to make their own situation worse, the methodology leaves much to be desired. When providing data, potentially conflicting interests among various stakeholders – supranational and international institutions, agencies, civil society, national authorities, networks of contact points, etc. – remain likely to lead to conflicting information on which the Commission would then have difficulty basing its assessment. When data is organized around indicators, systemic ROL breaches may also be harder to expose. The fact that all this data and information is concentrated in the hands of the one institution does not help in terms of transparency or checks and balances either, not to mention the fact that the same institution holds most prerogatives in ROL matters.

Two: apart from the Council’s annual rule of law dialogue and preliminary reference, the Commission is omnipresent in ROL mechanisms’ implementation. This seems to be consistent with the Commission being the “guardian of the treaties.”⁴⁷ But does the Commission really fulfil its role? Regarding Article 7 TEU, the fact that it was not the Commission but the European Parliament that exercised its right of initiative to trigger proceedings against Hungary⁴⁸ already points in the direction of what follows. Focusing on the right of initiative, the Commission holds a de facto exclusive one in infringement proceedings: in light of the fact that systemic ROL breaches have been constant in Hungary since 2010, it is astonishing that the Commission has only brought a total of eight infringement proceedings with a rule of law dimension against the MS concerned,⁴⁹ which sits quite far from risk-taking. As for conditionality mechanisms, both data problems and relative apathy give rise to concerns about the Commission potentially overshadowing the Council when it comes to exercising their joint assessment competence⁵⁰ or the Council’s deliberative competence reduced in practice to approving the Commission’s assessment in its proposal preparing the Council’s implementing decision.⁵¹ These examples clearly show that the Commission’s proactivity, or rather lack thereof,⁵² is a decisive factor in ROL mechanisms’ implementation.

Three: having a great number of institutions involved in implementing a mechanism seems to have a negative impact on its operability. Evaluation mechanisms are easy to implement because only one institution is involved. The case of conditionality mechanisms is similar: even though three institutions are involved, these mechanisms’ operability depends mainly on how the Commission exercises its prerogatives, and only marginally on the degree of uniformity of MS’ positions within the Council; the obligation to inform the European Parliament weighs close to nothing. However, reaction mechanisms are a lot more heterogeneous: all EU institutions as well as MS are involved in the implementation of at least one of them. Next to preliminary references, once infringement proceedings and actions for annulment are past the stage of being “slowed down” by the Commission, the CJEU’s deliberation takes place with the usual level of operability of judicial remedies. Then comes a contrast: both preventing and sanctioning arms of Article 7 TEU comprise four stages and involve up to three or four institutions. This requires greater cooperation and a more uniform interinstitutional consensus,⁵³ especially if all eight stages of the procedure are to be completed, not to mention that a considerable proportion of MS must also be in favour of the process.⁵⁴ On the plus side, the European Council’s involvement in the sanctioning arm of Article 7 TEU is of symbolic political significance.⁵⁵ However, when the Article 7 TEU procedure moves from the prevention phase to the sanction phase, the EP loses its right of initiative to be left only with a power of approval; this anticipates comments on the EU’s democratic deficit. It is not surprising that a high level of protection of constitutional values involves the largest number of institutions in rule of law mechanisms’ implementation. Nonetheless, high complexity in a decision-making process is undoubtedly another hindering factor to mechanisms’ operability.

Four: the European Parliament is virtually invisible in ROL mechanisms’ implementation. Even so, the EP still provides its own analyses on ROL issues in MS and invites other institutions to act.⁵⁶ The choice to leave parliamentary debate out of ROL mechanisms’ implementation could be justified by the need of executive solutions in crisis management, but not without inflating the EU’s democratic deficit.⁵⁷ Indeed, the quintessential institution of representative democracy⁵⁸ expressing the will of its citizens is only marginally called upon when a fundamental value of the EU is at stake. This does send a message.

To recap, ROL mechanisms’ operability is found to be reduced by issues 2 and 3: the low proactivity of the omnipresent Commission, on the one hand, and the number of institutions involved in a highly complex decision-making process, on the other hand. As for issues 1 and 4, data problems and democratic deficit raise questions that relate more to the mechanisms’ legitimacy, but do not seem to affect their operability.

The last step is to assess the performance of each ROL mechanism considering advantages and disadvantages deducted from the issues explored above. These factors make the achievement of the mechanisms’ predefined goals certain (“proven”), likely (“fair”), possible (“moderate”) or unlikely (“doubtful”). Table 3 awards these performance labels.⁵⁹

Table 3. Performance of rule of law mechanisms

In its current form, the Council’s annual rule of law dialogue is characterised by a “doubtful” performance due to the poorly developed procedural rules and the unlikeliness of MS’ cooperation. The same label is awarded to Article 7 TEU procedures: despite a symbolic and (theoretically) powerful nature, the number of institutions involved and internal modi operandi lacking effective political will considerably slow down the implementation process.⁶⁰

The remaining three evaluation mechanisms are placed in the “moderate” performance category because of the Commission’s margin of discretion and data problems. Particularly when facing a profile like Hungary’s, it is quite unfortunate that assessment methodologies allow for ROL breaches to be detected in specific areas but do not necessarily provide a systemic view,⁶¹ or that the Rule of Law Framework, aimed precisely at systemic violations, has not been applied to Hungary. The same label is assigned to the infringement procedure, whose advantages and disadvantages are of greater amplitude, but where the benefits of a judicial remedy are outweighed, again, by the Commission’s relative inaction, as well as by the fact that Hungary might simply ignore the CJEU’s judgments.⁶²

The advantages of a judicial remedy make action of annulment and preliminary reference the only mechanisms to obtain the “proven” performance label. However, this remains limited to actions brought before and declared admissible by the Court, leaving a number of sets of circumstances unable to produce even a merely indirect contribution to ROL’s protection.

Given that their implementation has not yet been completed, conditionality mechanisms’ performance is “fair” in the sense that the EU has adopted measures that suspend or withhold funding until Hungary implements the reforms that are necessary for the payments to be released.⁶³ Along these lines, the Commission, and later the Council, decided to release some of the retained funds considering that Hungary had performed sufficiently on some accounts.⁶⁴ As for the short-term follow-up of this end of 2023 decision, the Hungarian Prime Minister did not veto the aid to Ukraine for the first time in early February,⁶⁵ so the customary tug of war⁶⁶ might continue with a slightly bit more cooperation from Hungary’s part. However, given some features of the Hungarian legal system such as governance in multiple states of exception and decreased separation of powers,⁶⁷ it is not inconceivable for implemented reforms to be revoked or overridden soon after the funds’ collection. Should this hypothesis materialise, the performance label would of course have to be revised.

These considerations become even more eloquent when cross-analysing predefined goals and performance labels in Table 4.

Table 4. Intersection of predefined goals and performance labels

Predefined goals are ranked according to the intensity of their potential impact directly on the MS concerned. At one extremity, conceptual development of the ROL undoubtedly benefits the evolution of EU law as theorised, but it has the least direct impact on MS’ attitude in practice, as legal actions concern either the validity of acts or the interpretation of concepts. At the other extremity, suspension of a MS’s voting rights in the Council and its other rights while all other Treaty obligations continue to be binding on the same MS, followed directly by interim and financial measures that can be imposed in the context of (accelerated) infringement procedures, certainly have the potential to hurt a lot. In between, the intensity of the impact on MS varies from discussions with almost no consequences, through non-binding suggestions, to conclusions with consequences.

Overall, in light of the performance labels associated with the predefined goals, ROL mechanisms are only moderately efficient. Therefore, it must be concluded the EU’s ROL-related crisis management presents limited efficiency itself.

IV. Concluding remarks: The day can still be saved

In theory, the EU has two ways of dealing with undesirable events in the context of the rule of law crisis. Events that have not yet occurred do not really appear on the radar; the answer to the question of whether risk prevention policy is efficient is negative. Events that have already occurred do appear on the radar of the rule of law mechanisms; again, the verdict is no better than moderate efficiency. Since most undesirable events have already happened, developing a risk prevention policy at this point would probably be of theoretical interest only. That leaves either the adoption of new (kinds of) mechanisms, in a radical case accompanied by an abandonment of “old” ones – subject to subsequent reflection –, or the improvement the efficiency of the EU’s current crisis management policy.

If the aim is to keep even those with doubtful performance, upgrading the existing rule of law mechanisms could begin with a reflection on the Council’s annual rule of law dialogue. The choice to discuss rule of law-related issues in the Council’s General Affairs configuration could be interpreted as conferring a certain degree of importance to the matter.⁶⁸ However, to avoid giving the impression of a solely cosmetic solution by attributing an increasingly pressing issue to a catch-all configuration, the Council could also establish a monitoring agenda with precise indications as to how often and to what extent rule of law issues would be discussed, create an additional configuration or imagine a special mandate dedicated to this purpose. As for the other poorly performing mechanism, a reform of Article 7 TEU was initiated by a proposal of the European Parliament aiming to lower the required majorities, to establish a procedural agenda, to include the suspension of the right of the MS concerned to hold the Presidency of the Council, and to involve the CJEU as an arbitrary. Even though the first three amendments would be more than welcome, the choice to make the European Council exit and the Court of Justice enter the procedure leaves the same number of intervening institutions and only slightly reduces the procedure’s institutional complexity. That is, if the proposal ever makes it through Article 48 TEU’s ordinary revision procedure. At this point, it would also be worth considering whether to completely remodel Article 7 TEU or to set it aside altogether.

Turning to the Commission’s mechanisms assessing the state of the ROL in MS, it is not even methodological concerns that are the most embarrassing, but the absence of a legal basis: the EU Justice Scoreboard, the Rule of Law Review Cycle and the Rule of Law Framework are all based on non-binding Commission communications. If amending the Treaties were to prove unrealistic, the flexibility clause could come into play.⁶⁹ Indeed, the rule of law is an area subject to harmonisation (see conditionality mechanisms) and, despite Article 7 TEU’s theoretical potential to achieve the objective of promoting EU values, an argument could be put forward to the effect that the already existing provision provides for a power of action solely to punish, not to assess the rule of law. If such “sub-objectives” could not be distinguished under the conditions to activate Article 352 TFEU, the high degree of inoperability of Article 7 TEU and its doubtful performance could justify the use of the flexibility clause. If the CJEU were to approve this creative project similarly to selling arrangements,⁷⁰ as of today, the Council would still have to act unanimously, in agreement of MS that have no desire to be subject to rule of law assessment; to change this is also a part of the recent EP proposal.⁷¹ Alternatively, evaluation mechanisms could be linked to conditionality mechanisms, whether in the general regime of conditionality itself or in a related regulation, which would probably cause a loss on non-financial fronts, but still solve the problem. As it is, legality is as much a problem as the EU’s democratic deficit. Therefore, the two could be connected in the EP also being involved in assessments and taking some of the burden off the Commission; not only because the EP is so keen on hearing its own voice, but because a small democratic headwind might be healthy to counterbalance the Commission’s executive omnipresence.

One way of encouraging the Commission to act through infringement procedures would be to bring action for failure to act pursuant to Article 265 TFEU, as the EP did to speed up the implementation of Regulation 2020/2092.⁷² Another way would be to split the Commission’s approach in two by drawing a clear distinction between “general” failures to fulfil obligations under the Treaties and those relating to Article 2 TEU values⁷³: for the former, the Commission could maintain its selective attitude (or discretion) in order to preserve its reputation for strictness, whereas for the latter, a less cautious or even “automatised” referral to the CJEU could be imagined. To avoid Member States ignoring measures and judgments of the Court, the number and severity of pending and terminated infringement procedures could, again, be linked to conditionality mechanisms and modulate the release of funds by a simple amendment of relevant regulations.

It is time to abandon the separation of rule of law mechanisms based on their different approaches. Instead, links such as those mentioned above would allow to move beyond disorganised crisis management and towards a combined, strategical and systemic use of rule of law mechanisms.