Hostname: page-component-6898cfcb55-cbk24 Total loading time: 0 Render date: 2026-11-11T04:10:49.904Z Has data issue: false hasContentIssue false
  • English
  • Français

The Commission’s Power to Withdraw Legislative Proposals and its ‘Parliamentarisation’, Between Technical and Political Grounds

Published online by Cambridge University Press:  04 July 2018

Nicola Lupo
Rights & Permissions [Opens in a new window]

Abstract

European Commission – Monopoly on legislative initiative – Power to withdraw legislative proposals, as recognised and limited by the Court of Justice – Interinstitutional Agreement on Better Law-Making – European Parliament Rules of Procedure – Interinstitutional programming procedures – Technical and political usages of the withdrawal – Question of confidence in national parliamentary systems – National Parliaments and Early Warning System – Parliamentarisation of the decision to withdraw – Principle of institutional balance

Information

Type
Articles
Information
European Constitutional Law Review , Volume 14 , Issue 2 , June 2018, pp. 311 - 331
Copyright
© The Authors 2018 

Introduction

The Commission’s so-called monopoly on legislative initiative is manifestly one of the constant features of the EU legal order, mitigated only slightly by the attribution of such power in certain very closely circumscribed circumstances to other institutions and cases of ‘pre-initiative’ (for the European Parliament, the Council and citizens themselves).Footnote 1 Indeed, this power has significantly evolved as a consequence of the extension of EU policies and of the changing institutional setting. Political science studies converge in remarking that, especially after the entry into force of the Treaty of Lisbon, the Commission has often been influenced in exercising this power by the European Council’s conclusions and, at the same time, has become more cautious in so far as it tries to obtain prior consent from the European Parliament and/or the Council to potential initiatives.Footnote 2 All this, together with the already accomplished implementation of the internal market and the setting up of an ad hoc policy aimed at reducing legislative burden, has led to a significant decrease in the number of legislative proposals.Footnote 3

Nonetheless, there is no agreement as to whether this trend should be regarded as permanent or contingent, since much depends on which preferences any given scholar has for current or future EU institutional architecture and also on the foreseen function of EU law.Footnote 4 More in general, the role of the Commission remains after all somewhat undefined, somewhere between political leadership and policy management.Footnote 5 The Spitzenkandidaten experiment in the 2014 European Parliament elections seems understandably to have had some elements that point in the first direction,Footnote 6 although it is still not clear whether it could transform itself into a proper and stable constitutional convention.Footnote 7 At the same time, the increased role of the European Council, especially in response to the financial and migration crises, seems to be pressing on in the other direction.Footnote 8

One of the more recent developments has been the clear recognition by the Court of Justice of the Commission’s power, as a part of its power of legislative initiative, to withdraw legislative proposals provided that this is done within a time limit (‘as long as the Council has not acted’) and in compliance with the principles of conferral of powers, institutional balance and sincere cooperation. The judgment in question, delivered in April 2015, was – significantly – followed by provisions explicitly devoted to the power of withdrawal both in the new version of the Interinstitutional Agreement on Better Law-Making, of 13 April 2016, and in the European Parliament rules of procedure, as reformed on 13 December 2016. These follow-up provisions both seek to incorporate the decision to withdraw a legislative proposal into the interinstitutional programming procedures and, by this means, to anticipate and ‘parliamentarise’ it, wherever possible, thereby making the dynamics of Parliament and Commission more like those typical of parliamentary systems.

In order to place these new elements within the evolving EU institutional system, the analysis in this contribution proceeds from a recap of the monopoly of the Commission’s legislative initiative. It then goes on to address the contents of the judgment of the Court of Justice and its follow-ups, while discussing, in particular, the similarities with other procedural instruments by which Executives can influence the legislative process in some member states’ parliamentary forms of government. Finally, it seeks to understand the reasons that might lie behind the action of the Council before the Court of Justice and those that led the Court of Justice, in the current institutional setting, to decide in favour of the Commission. It concludes that the Court of Justice has protected the ‘political’ role of the Commission in the legislative process, avoiding the Commission’s confinement to being simply an ‘honest broker’, while at the same time seeking to ‘parliamentarise’ the exercise of the power to withdraw legislative proposals.

The Commission’s ‘monopoly’ on legislative initiative

It is well known that the European Commission holds a so-called ‘monopoly’ on legislative initiative in the EU. This means that, according to Article 17(2) TEU, ‘Union legislative acts may only be adopted on the basis of a Commission proposal, except where the Treaties provide otherwise’.Footnote 9 The exceptions, although progressively expanding, are still rather limited.Footnote 10 However, because of them, some authors prefer to speak of a ‘near-monopoly’, a ‘quasi monopoly’, a ‘quasi-exclusive right’, or similar expressions.Footnote 11

This monopoly or near-monopoly has functioned as one of the blueprints of the so-called ‘community method’ since its inception. The right to put forward formal proposals for legislative acts was not assigned to the intergovernmental body, the Council (of Ministers), which originally exercised the legislative function; neither was it attributed to the European Parliament, which has since acquired direct popular legitimacy and gradually seen its legislative powers increased. It was instead reserved for a supranational institution, not controlled by any Government, ‘completely independent’, called upon to ‘promote the general interest of the Union’ (Article 17, paragraphs 3 and 1, respectively, TEU).

This is attributable to the founders’ intention not to give advantage, in the drafting process of EU legislation, to any Member State, nor to a political party. The Commission ‘was supposed to be able to adopt legislative proposals that would be based on the most advanced national legislation or on innovative regulation that pursued the interest of the entire Community/Union’.Footnote 12 At the same time, the legislative initiative allows the Commission to make the crucial choice about which legal basis is to be used in order to adopt each legislative proposal: a choice that, as is well known, has many implications, first and foremost regarding the typology of the act to be adopted, the procedure to be followed and the majority required in the Council.Footnote 13 Finally, this setting was also fully consistent with the generally high level of technical complexity of the legislative texts to be drafted – which often went together with a low level of political salience – and with the need properly to take care of their multilingualism. In substantive terms, this made the Commission the institution to which all representatives of interest groups needed, and still need, to address, in order to have their voices heard from the start of the process that leads to the enactment of a legislative act.Footnote 14

Moreover, the importance of the legislative initiative derives from the traditional rule, reaffirmed by Article 293(1) TFEU, according to which the Council may normally only amend a proposal of the Commission by unanimous vote.Footnote 15 Consequently, the Commission has the advantage of determining the initial option, but this is not all: there is, indeed, a strong disincentive to change it, as, in absence of approval by the Commission, any amendment can be avoided if even a single representative of a member state sticks to the original text.

In practice, this Treaty rule means that ‘the Council votes only when the Commission has clearly adopted a position on any amendment of its proposal’ and the Commission ‘may amend its proposal in line with the Council’s desired amendments even orally during the meeting’.Footnote 16 The compliance with the Treaties of this institutional practice has been repeatedly confirmed by the Court of Justice, which recently denied the need for unanimity in the Council when ‘the amended proposal was approved on behalf of the Commission by two of its Members, who were authorised by the College of Commissioners to adopt the amendments concerned’.Footnote 17

For interest groups, this arrangement means that once one of them manages to get a provision inserted into a legislative proposal submitted by the Commission, the chances of its reaching the final stage of the long and complex legislative process are usually rather high.

The power to withdraw legislative proposals, as recognised (and limited) by the Court of Justice

The role of the Commission in the legislative process does not end with the formalisation and submission of the proposal. In addition to the power of legislative initiative stricto sensu (i.e. the right to trigger the legislative process by drafting and submitting a proposal), the Commission, in the course of the legislative process, is entitled both to amend/modify the proposal and to withdraw it.Footnote 18 However, while the former is expressly attributed by the treaties, ‘as long as the Council has not acted’ (Article 293(2) TFEU), the latter is not.

This is the main reason of interest of judgment 14 April 2015 of the Court of Justice of the EU (Grand Chamber), Case C-409/13 Council v Commission, which clearly recognised that the Commission has the power of withdrawal, identifying as its ‘constitutional basis’Footnote 19 Article 17(2) TEU in conjunction with Articles 289 and 293 TFEU. It is true that the power of withdrawal had been already recognised in a couple of previous judgments (ECJ 14 July 1988, Case 188/85, Fediol v Commission; ECJ 5 October 1994, Case C-280/93, Germany v Council). However, in none of these cases was the statement on the existence of this power formulated in such general terms as in the most recent judgment, so in both the legitimacy of the withdrawal could be derived by the specific characteristics of the act under consideration.Footnote 20

This explains why, until recently, some scholars still argued that the Commission was not entitled to withdraw its proposals, especially once the proposal had started to be discussed by the other institutions.Footnote 21 Also, the Council and the European Parliament doubted whether the Commission had such a power, primarily where its exercise was motivated by political reasons.

The identification of the constitutional basis of the power of withdrawal as Article 17(2) TEU, ‘read in conjunction with’ Articles 289 and 293 TFEU, means that this power is seen more as a natural corollary to the Commission’s power of legislative initiative and its legislative role, rather than as a mere aspect of the power to amend its proposals. In this, the decision seems to take some distance from the legal basis invoked by the Commission in the act at stake (that referred only to Article 293(2) TFEU), and follows the interpretation adopted by the Advocate General.Footnote 22 This approach appears correct, as the legal and political effects of the withdrawal of a legislative proposal on the two institutions co-exercising the legislative function are much more important than those deriving from one or more amendments proposed by the Commission.

Together with this general recognition of the power of withdrawal, the Court of Justice pointed to some constraints that the Commission must respect. In fact, the crucial assumption is that a general power of veto in the conduct of the legislative process attributed once and for all to the Commission ‘would be contrary to the principles of conferral of powers and institutional balance’.Footnote 23 This means that the Commission must state to the Parliament and the Council the ‘grounds for withdrawal’, that a time limit is established and that an action for annulment may be brought against the withdrawal before the Court of Justice, also in order to verify whether the decision is supported by cogent evidence or arguments.

This form of judicial review of the act of withdrawal was exercised in the case at issue, concerning a proposal for a regulation laying down general provisions for macro-financial assistance to eligible third countries and territories.Footnote 24 The Court held that the grounds for withdrawal were sufficiently brought to the attention of the Parliament and the Council and that they were capable of justifying the withdrawal. Indeed, the Court agreed with the Commission that the amendments the Parliament and the Council were planning to make, requiring the use of ordinary legislative procedure instead of an implementing act for granting macro-financial assistance to one or more third countries, would have prevented the achievement of the objectives pursued by the legislative proposal – which was intended to create a framework regulation in order to accelerate decision-making in that regard – and therefore deprived it of its raison d’être. Furthermore, the Court analysed in some depth the legislative process as it developed, and more specifically the outcomes of the ‘tripartite meetings’ (better known as ‘trilogues’).Footnote 25 It concluded that the Commission had not infringed the principle of sincere cooperation in that context, since it had even given prior notice of its intention to withdraw the proposal in case the Council and the Parliament did not reach an agreement on a different option.

In contrast, not much attention was given to the argument alleging an infringement of the principle of democracy (Article 10 TEU), as raised especially by the German government. In rejecting it, the Court recalled only that the power of the Commission to withdraw a proposal is inseparable from the right of initiative, without expanding on the question whether both powers, and its reservation in favour of the Commission, might be criticised or even assessed from a democratic perspective.Footnote 26 This is not a minor issue, indeed, at least from a constitutional viewpoint, especially as both these powers can have significant legal and political effect on institutions that have a stronger democratic legitimacy than the Commission, namely the Council and, most of all, the Parliament (which, as is well-known, is deprived of any right of legislative initiative), between which the judgment does not draw any distinction.Footnote 27 It must be noted, however, that more attention has been given to this kind of argument in two more recent judgments of the General Court on the citizens’ initiative, in order to justify a use of that new instrument which is significantly wider compared to that allowed by the Commission, which is naturally inclined to be overprotective of its monopoly on legislative initiative.Footnote 28 It remains to be seen whether these decisions will open a new season in the way in which the case law of the Court of Justice deals with the principle of democracy.Footnote 29

Two follow-ups in the parliamentary process: in the Interinstitutional Agreement on Better Law-Making and the new Rules of Procedure of the European Parliament

Principles and limits to the Commission’s power of withdrawal as defined by the Court of Justice in Case C-409/13 have been promptly implemented in the months that followed the judgment, first by the Interinstitutional Agreement on Better Law-Making between the European Parliament, the Council of the European Union and the European Commission of 13 April 2016;Footnote 30 second, by the European Parliament rules of procedure, as amended on 13 December 2016.Footnote 31 In both cases, some specific provisions on the power of withdrawal have been inserted, with the intent of encouraging a process of proceduralisation and parliamentarisation of its exercise.

More specifically, point 9 of the new Interinstitutional Agreement deals with the power of withdrawal and echoes the wording of the judgment. It requires the Commission to give a proper statement of reasons and to indicate its future intentions:

‘In accordance with the principles of sincere cooperation and of institutional balance, when the Commission intends to withdraw a legislative proposal, whether or not such withdrawal is to be followed by a revised proposal, it will provide the reasons for such withdrawal, and, if applicable, an indication of the intended subsequent steps along with a precise timetable, and will conduct proper interinstitutional consultations on that basis. The Commission will take due account of, and respond to, the co-legislators’ positions (emphasis added)’.

As has been remarked, in this way the right of initiative ‘is maintained but under the shadow of a renewed burden of political justification’.Footnote 32

Also, point 8 of the Interinstitutional Agreement refers to the withdrawal of legislative proposals, in connection with the Commission Work Programme. According to the Agreement, this document, in fact, is called upon to include ‘major legislative and non-legislative proposals for the following year, including repeals, recasts, simplifications and withdrawals’ (emphasis added).

Of this provision, where it refers to withdrawals, a twofold interpretation could be given. On the one hand, it reaffirms that the Commission is called, annually, to review the status of its proposals, verifying whether it is appropriate to withdraw those that did not receive any attention either from the Parliament or from the Council, and those that can be deemed obsolete: the so called ‘technical’ or ‘administrative’ withdrawals, largely used especially by the Barroso and Juncker Commissions, also in order to achieve better-regulation objectives.Footnote 33 On the other hand, it could also be read as intended to reflect a duty, recognised by the Commission itself, to pre-alert the other institutions of its intentions in the context of the interinstitutional programming procedures: i.e. to notify in advance, at least in the normal course of events, its ‘political’ decision to withdraw a proposal through the insertion of this intention in the Commission Work Programme.

Consistently with the latter interpretation, in the framework of a broader revision of the European Parliament Rules of Procedure, a new provision regarding the power of withdrawal has been situated in Rule 37, relating to ‘annual programming’, even though that rule sets out a procedure which would seem to be applicable also beyond the scope of the examination of programming documents. Rule 37(4) specifies that the intention to withdraw a legislative proposal needs to be discussed in the competent parliamentary committee and, where appropriate, also in the plenary:

‘If the Commission intends to withdraw a proposal, the competent Commissioner shall be invited by the committee responsible to a meeting to discuss that intention. The Presidency of the Council may also be invited to such meeting. If the committee responsible disagrees with the intended withdrawal, it may request that the Commission make a statement to Parliament’ (emphasis added).Footnote 34

In all these follow-ups of the judgment, the intention to ‘proceduralise’ and even to ‘parliamentarise’ the power of withdrawal, by repeating, implementing and – up to a certain extent – even expanding the limits to this power as defined by the Court of Justice, emerges. It is clear that the recognition of this power means that the Commission has a mighty weapon, as it can be used throughout the first steps of the legislative process to impede any amendment the Commission does not like. However, it is a weapon that should be used loyally and, most of all, in a timely manner, avoiding surprises: so that both the Parliament and the Council may exercise their legislative function without the unexpected risk that all the work already done, until the first reading by the Council, could be nullified by an unexpected and unreasoned withdrawal of the proposal by the Commission.

This means that the Commission, if it has already decided to withdraw a given proposal, should communicate its decision in the programming procedures, independently from the technical or political grounds invoked.Footnote 35 As the programming procedures are now designed as fully interinstitutional in order to diminish conflicts among (EU and even national) institutions in the following steps of the decision-making, this inclusion determines, with regard to the power of withdrawal, a proceduralisation of a power that the Court did not want to leave exclusively to the Commission’s discretion.

In order to fully understand the need for this proceduralisation, it must be recalled that the concrete practice of the ordinary legislative procedure now tends to concentrate all the negotiations in the first reading, thanks to the intense use of the aforementioned ‘trilogues’ and early agreements.Footnote 36 This has a positive effect in terms of the timeframe of completion of the process, but at the same time it tends to anticipate the negotiations, which often take place even before the actual start of the examination by Parliament’s committees.

This implies that the time limit established for the Commission’s amendments, according to which they could be submitted ‘as long as the Council has not acted’ (Article 293(2) TFEU), and now applied by the Court of Justice also to the power of withdrawal, has changed in significance. The idea was that, once the Council ‘has acted’ – in the ordinary legislative procedure, either approving the European Parliament’s position, according to Article 294(4) TFEU, or adopting its position at first reading, according to Article 294(5) TFEU – the Commission was forbidden to make use of its power of amendment.Footnote 37

However, today the ordinary legislative procedure ‘has de facto become a single-reading legislative procedure’.Footnote 38 That is why both the European Parliament rules of procedure and the interinstitutional agreement on better law-making aim at anticipating the Commission’s decision to withdraw so as to avoid a waste of time and resources especially on the part of the Parliament, which would be inevitable if withdrawals were to take place, unexpectedly, after some parliamentary examination and votes had already developed.

A parallel with the question of confidence?

In the judgment, and more clearly in the Opinion of the Advocate General, the Commission’s power of withdrawal is contrasted with another means of interrupting the legislative process: the power of veto, which is generally conferred on the head of state ‘in order to prevent, with suspensory or definitive effect, the entry into force’ of a piece of legislation already approved by Parliament.Footnote 39

In the EU, the position of the Commission changes during the legislative process. For instance, it may only may amend its proposal ‘as long as the Council has not acted’, according Article 293(2) TFEU. The same time limit should be applied to the power of withdrawal according to the Court and the Advocate-General. That the power of withdrawal may be used only ‘as long as the Council has not acted’, while vetoes are exercised at the end of the legislative process (within or outside it, depending on the relevant constitutional system), on the final legislative text, indeed means that the power of withdrawal always stays well away from an ex post veto. There are other differences too. Vetoes are, after all, typically employed by heads of state, monocratic bodies with a higher formal status within the institutional hierarchy, two features the Commission has not.

Be that as it may, based on the procedural and temporal features as defined by the interinstitutional agreement and the European Parliament Rules of Procedure, the Commission’s power of withdrawal exhibits certain similarities with another instrument, rather common at the national level in parliamentary forms of government: the ‘question of confidence’, through which the executive is allowed – implicitly or even explicitly – to link ex ante its destiny to the outcome of a parliamentary vote, often during the legislative process.Footnote 40 According to the comparative analysis, the ‘question of confidence’ is the confidence procedure used, generally upon initiative of the Prime Minister, ‘after government formation is complete and in the context of legislative debates on specific policy issues or specific aspects of the government’s program’.Footnote 41 It aims at complementing, as an instrument of last resort, the crucial role that Executives regularly play throughout the legislative process, in every parliamentary form of government.

Of course, also within this parallel, many distinctions need to be drawn, especially because of the ultimate legal consequences: while with posing the question of confidence the Executive threatens to resign if the relevant proposal is not adopted, the ultimate consequence of the power of withdrawal is clearly less traumatic, consisting of the forced termination of the specific legislative process which has been started. However, two elements must be considered. First, the question of confidence, in some jurisdictions, is used so frequently, and even independently of the actual importance of the legislative proposal at stake within the government’s overall programme, that it becomes an ‘ordinary’ procedural tool. In these cases, the threat of the government’s resignation is mainly theoretical and almost never materialises in practice, because the parliamentary majority almost always rallies behind the Government.Footnote 42 Second, and most importantly, together with the question of confidence, several constitutions or parliamentary rules of procedure provide for a series of other procedural instruments that can be used by the Executive in order to influence the legislative process, especially concerning forms of blocked vote or the rejection of amendments that would not be agreed by the Executive itself.Footnote 43

Furthermore, an element common to the withdrawal of legislative initiatives and the question of confidence can be spotted. Through both these instruments, the government or the Commission are declaring that they do not agree with a possible modification of a legislative bill and are stating in advance that if modification takes place anyway, they will make use of their respective weapons (clearly a mightier weapon in the case of the question of confidence). Therefore, both these declarations share a common aim: to influence subsequent steps of the legislative process in order to get it closer to Executives’ own position.Footnote 44

Technical and political uses of the power of withdrawal

Obviously, the parallel between the power of withdrawal and the question of confidence cannot be generalised. It could apply, with the said caveats, only to cases in which the decision to withdraw a legislative act has a dominant political objective: it is aimed, thus, at impeding the approval of a text with which the Commission disagrees on the ground that its nature has changed in comparison with the original proposal (this is referred to in French as ‘dénaturation’).

Different treatment could possibly be imagined for cases in which a decision to withdraw is taken for technical or administrative reasons, e.g. because the legislative proposal has not proceeded any further; has become obsolete due to the emergence of new circumstances or of technical or scientific data; has been rendered clearly ineffective or has at least lost its topicality in relation to new events which have supervened in law or in practice (or else because either the Parliament or the Council definitely did not want it). In sum: when the withdrawal is restricted to objective circumstances, independent of the Commission’s specific interest.

In these cases, the parallel in the national legal orders should also probably be different; the instrument to be considered should this time be identified, generally speaking, as the government’s power to withdraw its legislative bills. Such a power is indeed often recognised to the government, as a sort of corollary to the power of initiative,Footnote 45 but its effects are not as relevant: ‘parliament may continue to examine a bill withdrawn by its original sponsors if some other holder of the power of initiative adopts it or if the house itself decides to proceed of its own initiative’.Footnote 46 This is something that evidently, because of the Commission’s monopoly on legislative initiative, cannot take place in the EU legal order.

In the first decades of the EU legal order, the power of withdrawal was used by the Commission almost exclusively for technical reasons.Footnote 47 Indeed, this kind of withdrawal did not raise much concern. Nor did much controversy arise when the second Barroso Commission (especially since 2012) and the Juncker Commission (since 2014) decided, also under the REFIT (Regulatory Fitness and Performance) programme, to withdraw a certain number of pending proposals in order to simplify EU law and to cut through (potentially new) red tape.Footnote 48

Instead, the power of withdrawal for political reasons has traditionally been used with extreme caution, while it has increased more recently, mainly as a form of reaction to potential legislative agreements, at the expense of the Commission, between the Council and the Parliament. To put it differently, the power of withdrawal has been used, or was used as a threat by the Commission, whenever the legislature intended to amend a Commission proposal in a way that the Commission disliked for political reasons.

Indeed, the principles affirmed by the Court of Justice, and even the procedures determined by the follow-ups, make more sense precisely where there are political issues at stake. When this is the case, the grounds of withdrawal need to be clarified and discussed publicly and some degree of ‘parliamentarisation’ of the political and institutional conflict has to take place, consistent with the need to keep a high level of transparency of the legislative process (including its forced termination).

In other words, the controversy about the power of withdrawal seems to be linked to the increasingly political use of this instrument. The judgment and its follow-ups are fully aware of the multiple usage of this power. As politics, gradually, makes its way into the EU decision-making process, mingled with many other legislative initiatives that remain a predominantly technical, a distinction could have been made between decisions that have formally the same name, but are characterised by very different political and institutional weight. However, in the EU legal order, at least for the moment, the task of selecting which withdrawals have to be specifically discussed from a political perspective has been left to the institutions involved.

A rather peculiar withdrawal at the request of national parliaments

The EU legal order reserves a particularly peculiar treatment for the eventuality in which the Commission decides to withdraw a legislative proposal in the context of the procedure of the Early Warning System, as provided for in Protocol no. 2 annexed to the Treaty of Lisbon: i.e. after a yellow card or an orange card has been issued by national parliaments, signalling a violation of the principle of subsidiarity.Footnote 49 This has happened, until present, only once; on two other occasions, a yellow card was raised, but the Commission decided to maintain its proposals, giving reasons for those decisions.Footnote 50

It needs to be clarified, first, that in this case there could be no doubt, since the beginning, about the legality of the withdrawal. Here it is the Treaty of Lisbon itself, through one of its protocols, which explicitly provides for the option of withdrawal of a proposed legislative act by the Commission (or the other institution that issued the draft legislative act), as an alternative to amending or maintaining it, in any case giving reasons for the choice made.Footnote 51 It seems difficult, however, given the specificity of this procedure, to derive from the aforementioned provision any signal in favour of a general power to withdraw legislative proposals. In this regard, the decision of the Court of Justice departed rather clearly from the opinion of the Advocate General, who had argued in this sense,Footnote 52 and preferred not to refer to the Early Warning System, probably because that procedure may also be used, in an a contario reasoning, as an argument to deny the Commission a general power of withdrawal.

However, once the legality of the power of withdrawal has been recognised, on a general basis, by the Court of Justice, there seems to be nothing to prevent the Commission from using the reasoned opinions of national parliaments on compliance with the subsidiarity principle – also when they do not reach the thresholds established by Protocol No. 2 –, or perhaps even the contributions issued by national parliaments in the context of the political dialogue,Footnote 53 as grounds to justify a decision to withdraw a given proposal. Reasoned opinions as well as contributions submitted by national parliaments are part of the decision-making process, and all EU institutions, starting with the one that took the initiative, are called to consider them as they please in the following steps of the process (especially in the first reading).

Second, it is permissible to ask whether a withdrawal resulting from a yellow/orange card – or reasoned opinions from national parliaments insufficient in number to reach the threshold for a yellow card, or maybe even from remarks made during political dialogue – should be qualified as technical or as political. The task is not made easier by the fact that, by its very nature, the principle of subsidiarity is notoriously ambiguous and multipurposed, somewhere in between law and politics, legal analysis and political assessment.Footnote 54 Scholars who have analysed the Early Warning System are indeed divided, on the reading of this procedure and on the scope of parliamentary scrutiny.Footnote 55

Some argue that national parliaments are called upon to play the role of mere legal advisors on compliance with the principle of subsidiarity, which is to be given a strict interpretation.Footnote 56 In this way, these scholars probably sought to reduce the effects of their direct involvement in EU decision-making on the ground that national parliaments are heterogeneous and potentially dangerous when inserted into the logic of European integration.

Others maintain that, on the contrary, to better fulfil the function of ‘reconnecting’ EU democracy by bringing more national politics into largely Europeanised policies, it is preferable that the Early Warning System should be read and used as a political rather than as a merely legal instrument.Footnote 57 At least in the sense that each national parliament – and even each House, in bicameral systems – is relatively free to interpret the parameters of its own assessments and to determine the scope of its interventions. Consequently, it comes as no surprise that in some cases, they act almost like legal advisors, interpreting in a strict sense their role as guardians of the principle of subsidiarity; while in other cases, they use reasoned opinions based on the principle of subsidiarity as mere excuses for raising political objections to a given proposal.Footnote 58

Be that as it may, it seems that the decision to withdraw (or, equally, a decision to maintain the proposal or to amend it) has a clear political nature whenever it happens to be motivated by positions expressed by national parliaments, which are foremost political actors, even though the decision is based on a presumed violation of one of the principles affirmed by the Treaties.

It is interesting to note that until now the Commission has followed different approaches. In the case of the first yellow card (on a proposal for a Council regulation on the exercise of the right to take collective action: the so-called ‘Monti II’), for instance, it reaffirmed its position that the principle of subsidiarity had not been infringed, but decided in any case to withdraw its draft Regulation, recognising that it was ‘unlikely to gather the necessary political support within the European Parliament and the Council to enable its adoption’.Footnote 59 Contrarily, in the case of the third yellow card (on a proposal for a Directive of the European Parliament and of the Council concerning the posting of workers), it tried to conceal the political nature of its decision to keep the draft directive by issuing an ad hoc communication in which it did not consider at all the important political, social and geographical issues and rifts that the merit of the proposal raised.Footnote 60

Conclusion: inter-institutionalising the powers of initiative and withdrawal

The withdrawals within the Early Warning System remain qualitatively peculiar and quantitatively marginal. More in general, the main factor that can justify the recent judicialisation of the controversy on the power of withdrawal is linked to the evolution of the EU legislative process, and to the institutional setting of the EU.

The monopoly of legislative initiative was conceived, and the related power of withdrawal first exercised, in a phase in which the only legislator was the Council of Ministers and the agenda-setting power was exclusively in the hands of the Commission.

Since then, not only have the policies of the EU undergone quite dramatic expansion, but also the legislative process has gone through a long, important but often hidden evolutionary path, in which it has been profoundly transformed. On the one hand, as already remarked, the agenda-setting power is now, at least for a significant part, in the hands of the European Council. That institution, although formally devoid of any legislative power (Article 15(1) TEU), requests a significant amount of legislative initiative from the Commission.Footnote 61 On the other hand, the increase in the legislative power of the European Parliament and the development of the ordinary legislative procedure, especially with the diffusion of trilogues, have deeply changed the nature and the timing of the law-making process.

As has been observed, the transformations involved all three elements of legislative initiative as identified by former commissioner António Vitorino. According to him, the monopoly on the legislative initiative consisted of a threefold power: ‘agenda setting’, definition of the ‘terms of debate’, and coordination of the negotiations that lead to the finalisation of the texts.Footnote 62

The transformations that have taken place have gradually yet profoundly changed the role of the Commission in the legislative process. Especially when there is no input from the European Council, in submitting its legislative initiatives the Commission seems eager to obtain some kind of prior consent from the Parliament and the Council, so as to avoid standstill situations once the proposal has been submitted. At the same time, the ordinary legislative procedure and then the diffusion of early agreements have in fact ‘put the Commission in a situation of structural disadvantage compared to previous decision-making arrangements’, limiting its room of manoeuvre considerably.Footnote 63 Nowadays, ‘in short, the Commission has to navigate very narrow political straits’.Footnote 64

The decision of the Court of Justice to recognise that the Commission has the power of withdrawal could be read and justified in the light of this new institutional context.Footnote 65 It can be seen as a way of avoiding the confinement, even in the first steps of the legislative process, of the role of the Commission as ‘honest broker’, one clearly incompatible with a withdrawal exercised on political grounds; and as a means of bringing it closer, consistent with its recently increased politicisation, to the role played by Executive in parliamentary forms of government, one of the protagonists in the legislative process.

At the same time, once the Commission’s power of legislative initiative is significantly reduced and strongly self-constrained in favour of the other institutions, starting with the European Council, there should be no hesitation made in clearly recognising that the Commission also has the parallel power of withdrawal, provided that this power is also exercised allowing some degree of involvement for the other institutions, consistent with the principle of institutional balance, which is confirmed as the fundamental constitutional principle ruling EU decision-making. According to this logic, the attempt to ‘parliamentarise’ this power, inserting the decision to withdraw a legislative proposal into interinstitutional programming procedures, takes on a different complexion. In the EU institutional system, the risk of a standstill tends to be seen as more dangerous than an overlap of competences and the responsibility that interinstitutional procedures and negotiations inevitably bring with them.

Footnotes

*

Full Professor of Public Law and Director of the Centre for Parliamentary Studies LUISS University, Rome. The author would like to thank the reviewers as well as Robert Bray for their comments on previous versions of this article.

1 See Gianfrancesco, E., ‘Article 17’, in H.J. Blanke and S. Mangiameli (eds.), The Treaty on European Union (TEU). A Commentary (Springer 2013) p. 681 , specifically at p. 699 ff.

2 Inter alia, see Rasmussen, A., ‘Challenging the Commission’s right of initiative? Conditions for institutional change and stability’, 30(2) West European Politics (2007) p. 244 ; P. Ponzano et al., The Power of Initiative of the European Commission: A Progressive Erosion?, Notre Europe, no. 89, 2012, spec. 43 ff; Kreppel, A. and Oztas, B., ‘Leading the Band or Just Playing the Tune? Reassessing the Agenda-Setting Powers of the European Commission’, 50 Comparative Political Studies (2017) p. 1118 ff.

3 See A. Tajani et al., Activity Report on the Ordinary Legislative Procedure (4 July 2014 - 31 December 2016), PE 595.931, 2017, available at <www.epgencms.europarl.europa.eu/cmsdata/upload/7c368f56-983b-431e-a9fa-643d609f86b8/Activity-report-ordinary-legislative-procedure-2014-2016-en.pdf>, visited 18 April 2018.

4 On the alternative between positive/negative integration see, recently, Schuetze, R., From International to Federal Market. The Changing Structure of European Law (Oxford University Press 2017).

5 To use the title of Chang, M. and Monar, J. (eds.), The European Commission in the Post-Lisbon Era of Crises. Between Political Leadership and Policy Management (Peter Lang 2013). See also Jacqué, J.P., ‘Lost in Transition: The European Commission between Intergovernmentalism and Integration’, in D. Ritleng (ed.), Independence and Legitimacy in the Institutional System of the European Union (Oxford University Press 2016) p. 15 . On how the Commission perceives itself cf Kassim, H. et al., The European Commission of the Twenty-First Century (Oxford University Press 2013) p. 130 ff.

6 On the origins of the Spitzenkandidaten practice see Priestley, J. et al., The Making of a European President (Palgrave MacMillan 2015). On its first effects, with different approaches, Hobolt, S.B., ‘A Vote for the President? The Role of Spitzenkandidaten in the 2014 European Parliament Elections’, 10 Journal of European Public Policy (2014) p. 1528 ; Fabbrini, S., ‘The European Union and the Puzzle of Parliamentary Government’, 5 Journal of European Integration (2015) p. 571 ; Goldoni, M., ‘Politicising EU Lawmaking? The Spitzenkandidaten Experiment as a Cautionary Tale’, 3 European Law Journal (2016) p. 279 ; Christiansen, T., ‘After the Spitzenkandidaten: fundamental change in the EU’s political system?’, 39 West European Politics (2016) p. 992 ; Kassim, H., ‘What’s new? A first appraisal of the Juncker Commission’, 16 European Political Science (2017) p. 14 .

7 The Spitzenkandidaten procedure has been qualified, although critically, as a constitutional convention by P.W. Post, The Spitzenkandidaten Procedure. Genesis and Nemesis of a Constitutional Convention?, LLM Thesis, Leiden University, 2015, available at <njb.nl/Uploads/2015/9/LLM-Thesis---LLM-European-Law---Paul-W.-Post.pdf>, visited 18 April 2018). More recently, see the Commission Recommendation (EU) 2018/234 of 14 February 2018 on enhancing the European nature and efficient conduct of the 2019 elections to the European Parliament (OJ L 45, 17.2.2018, p. 40), much more explicit, on the point, than the previous one (Commission Recommendation 2013/142/EU of 12 March 2013 on enhancing the democratic and efficient conduct of the elections to the European Parliament (OJ L 79, 21.3.2013, p. 29)).

8 See, among many, Puetter, U., The European Council and the Council. New Intergovernmentalism and Institutional Change (Oxford University Press 2014).

9 On the contrary, in the case of non-legislative acts, the criterion is inverted, as, according to the following sentence, the Commission’s initiative needs to be specifically provided for by the Treaties.

10 For an accurate reconstruction of the exceptions, already existing in the Treaty of Rome and added with the Treaty of Maastricht and then with the Treaty of Lisbon inter alia in judicial cooperation in criminal matters and police cooperation (Art. 76 TFEU: allowing also a quarter of the Member States to initiate EU legislation), see Ponzano, supra n. 2, p. 8 and L. Guilloud-Colliat, ‘Rapport Union Européenne’, in L’initiative de la loi (ForInCip 2017) p. 181 ff.

11 See Konstadinides, T., Division of Powers in European Union Law: The Delimitation of Internal Competence Between the EU and the Member States (Kluwer 2009) p. 63 ; Costa, O. and Brack, N., How the EU Really Works (Ashgate 2014) p. 73 .

12 Ponzano, supra n. 2, p. 7.

13 On the choice of the legal basis as based on objective factors see Leino, P., ‘The Politics of Efficient Compromise in the Adoption of EU Legal Acts’, in M. Cremona and C. Kilpatrick (eds.), EU Legal Acts. Challenges and Transformations (Oxford University Press 2018) p. 30 at p. 35.

14 Bouwen, P., ‘The European Commission’, in D. Coen and J. Richardson (eds.), Lobbying the European Union. Institutions, Actors, and Issues (Oxford University Press 2009) p. 19 .

Furthermore, this Treaty provision is relevant also in drawing the distinction between items that can be included among those to be approved without discussion (A items) and those on which a discussion will take place (B items). According to Art. 3(8) of the Council rules of procedure, in fact, an ‘A’ item shall be withdrawn from the agenda, unless the Council decides otherwise, if […] a member of the Council or the Commission so requests’.

15 Lenaerts, K. and Van Nuffel, P., European Union Law (Sweet & Maxwell 2011) p. 660 (also for references to the limited exceptions to this rule). On possible ways of circumventing this rule through early agreements with the Parliament see Corbett, R. et al., The European Parliament (Harper 2016) p. 280 .

16 Comments on the Council’s Rules of Procedure, General Secretariat of the Council, Brussels, 2016, p. 53 ff, available at <www.consilium.europa.eu/media/29824/qc0415692enn.pdf>, visited 18 April 2018. For the case law of the Court of Justice see ECJ 5 October 1994, Case C-280/93, Germany v Council, para. 36.

17 See ECJ 6 September 2017, Joined Cases C-643/15 and C-647/15, Slovak Republic and Hungary v Council, paras. 177-189 (referring to the judgment in Case C-409/13), also quoting Art. 13 of the Commission’s Rules of Procedure: if ‘interpreted in the light of the objective of Article 293(2) TFEU of protecting the Commission’s power of initiative’, it follows ‘that the College of Commissioners may authorise one or more of its Members to amend, in the course of the procedure, the Commission’s proposal within the limits that the College has previously defined’.

18 See A. Rasmussen, supra n. 2, p. 246 ff (stressing the importance of the additional rights that ‘aim to protect the proposals throughout the policy process once introduced’).

19 The expression ‘constitutional basis’ is not used by the judgment, but appears in the Opinion of AG Jääskinen (delivered on 18 December 2014: points 36 and 52). According to this Opinion, moreover, the dispute itself is ‘constitutional in nature’ (point 1).

20 As remarked by Adam, R., ‘Il potere di iniziativa della Commissione e il processo decisionale: il difficile equilibrio tra ritiro della proposta e potere decisionale’ [The Commission’s Power of Initiative and Decision-making Process: The Difficult Balance Between Withdrawal and Decision Power], in A. Tizzano (ed.), Verso i 60 anni dai trattati di Roma. Stato e prospettive dell’Unione europea [Towards 60 Years of the Treaties of Rome: State and Perspectives of the EU] (Giappichelli 2016) p. 21 at p. 24.

21 Adam, R. and Tizzano, A., Manuale di diritto dell’Unione europea [Handbook of EU Law] (Giappichelli 2014) p. 200 ff, arguing that in this case the Commission would be guilty of a ‘sviamento di potere’ (détournement de pouvoir, usually translated as ‘misuse of power’).

22 See the Opinion of the Advocate General, stating that ‘the power of withdrawal stems from the role conferred on the Commission in the context of the legislative process […] not merely as the body which will in the future implement the legislative provisions to be adopted by the Parliament and the Council, but also as the custodian of the general interest of the European Union’.

23 ECJ 14 April 2015, Case C-409/13, Council v Commission, para. 75.

24 For the peculiarity of the case at stake see Ninatti, S., ‘Un conflitto tra Consiglio e Commissione: la conferma del ruolo della Corte come arbitro dell’equilibrio istituzionale nella forma di governo dell’Unione’, 35(1) Quaderni costituzionali (2015) p. 795 and Ritleng, D., ‘Does the European Court of Justice take democracy seriously? Some thoughts about the Macro-Financial Assistance case’, 53(1) Common Market Law Review (2016) p. 11 at p. 26 ff.

25 On ‘trilogues’ see, among many, Rasmussen, A. and Reh, C., ‘The consequence of concluding codecision early; trilogues and intra-institutional bargaining success’, 20(7) Journal of European Public Policy (2013) p. 1006 ; Roederer-Rynning, C. and Greenwood, J., ‘The culture of trilogues’, 22(8) Journal of European Public Policy (2015) p. 1148 ; Roederer-Rynning, C. and Greenwood, J., ‘The European Parliament as a developing legislature: coming of age in trilogues?’, 24(5) Journal of European Public Policy (2017) p. 735 .

26 For criticism of the monopoly on initiative deriving from a lack of democratic credentials see, inter alia, R Mastroianni, ‘L’iniziativa legislativa nel processo legislativo comunitario tra deficit democratico ed equilibrio interistituzionale’ [Legislative Initiative in European Communities’ Law-making Process between Democratic Deficit and Inter-institutional Balance], in S. Gambino (ed.), Costituzione italiana e diritto comunitario [The Italian Constitution and European Community Law] (Giuffrè 2002) p. 433 at p. 437 ff, and Martines, F., ‘Institutional Balance, Democracy and Agenda Setting in the European Union’, in L. Daniele et al. (eds.), Democracy in the EMU in the Aftermath of the Crisis (Giappichelli-Springer 2017) p. 141 .

27 This is the main criticism to the decision levelled by Ritleng, supra n. 24, p. 19 ff, according to whom the Court of Justice ‘seems to be largely underpinned by an out of date functionalist view of the European integration process’ that ‘does not take any account of the rise of the principle of democracy in the institutional system of the European Union’.

28 See judgments 3 February 2017 (Case T-646/13) and 10 May 2017 (Case T-754/14), in which the General Court annulled two decisions of the Commission not to register citizens’ initiatives. The second one seems more relevant here, not only because it was grounded on more explicit reasoning (e.g. on the principle of democracy as ‘one of the fundamental values of the European Union’ and, at the same time, as ‘the objective specifically pursued by the ECI mechanism’), but also because it referred to the Commission’s power of withdrawal and to amend its proposals (although regarding not a legislative act, but an authorisation to open a negotiation for an international agreement), using it somehow against the Commission itself. The Court stated, in fact, that the citizens’ initiative ‘includes the power to request an amendment of legal acts in force or their annulment, in whole or in part’ and that ‘therefore, nothing justifies excluding from democratic debate legal acts seeking the withdrawal of a decision authorising the opening of negotiations with a view to concluding an international agreement’. In September 2017 the Commission submitted a proposal for a regulation (COM(2017) 482 final) aimed at substituting the current regulation on the citizens’ initiative (Regulation (EU) No 211/2011 of the European Parliament and the Council of 16 February 2011). See Organ, J., ‘EU Citizens Participation, Openness and the European Citizens Initiative: the TTIP Legacy’, 54 Common Market Law Review (2017) p. 1713 .

29 For accurate analyses of the previous conceptions of the democratic principle see Leanerts, K., ‘The Principle of Democracy in the Case Law of the European Court of Justice’, 62 International & Comparative Law Quarterly (2013) p. 271 (arguing that these conceptions are not limited to the participation by the European Parliament in the legislative process, but also encompass other forms of governance) and S. Ninatti, Giudicare la democrazia? [Judging Democracy?] (Giuffrè 2004) p. 35 ff and 63 ff (showing how the institutional balance has been at the foundation of the democratic character of EU decision-making).

30 OJ L 123/1, 12.5.2016.

31 See European Parliament Rules of Procedure: general revision (2016/2114(REG)), at <www.europarl.europa.eu/sides/getDoc.do?pubRef=-//EP//NONSGML+TA+P8-TA-2016-0484+0+DOC+PDF+V0//EN>, visited 18 April 2018.

32 Dawson, M., ‘Better Regulation and the Future of EU Regulatory Law and Politics’, 53 Common Market Law Review (2016) p. 1209 . A reference to the power of withdrawal was already inserted within the Framework Agreement on relations between the European Parliament and the European Commission of 20 October 2010 (OJ L 304, 20.11.2010, p. 47), but the duty to give reasons was limited to the withdrawal of ‘any proposals on which Parliament has already expressed a position at first reading’ (point 39).

33 See D. Jančić, ‘The Juncker Commission’s Better Regulation Agenda and Its Impact on National Parliaments’, in C. Fasone et al. (eds.), Parliaments, Public Opinion and Parliamentary Elections in Europe (2015) 18 EUI Max Weber Working Paper, p. 45 ff, available at <cadmus.eui.eu/handle/1814/37462>, visited 18 April 2018) and, with interesting critical remarks, Dawson, supra n. 32, p. 1223 ff.

34 Rule 37(4) concludes by recalling that ‘Rule 123 shall apply’: that is, the procedure provided for statements from the Commission, Council and European Council, which allows the submission of and the vote on a resolution by the European Parliament, by which the Parliament can exercise its influence on how, when and even whether the Commission is going to make use of its power of withdrawal.

35 In the most recent Commission annual work programs, an ad hoc ‘annex 4’ contains the list of the pending legislative proposals which the Commission intends to withdraw within six months. See, for instance, the Commission Work Programme 2018. An agenda for a more united, stronger and more democratic Europe (COM(2017) 650 final, 24 October 2017). Of the reasons to justify the 15 intended withdrawals, obsolescence is the most frequent, although other grounds (including ‘no foreseeable agreement’) are also invoked. In this document, no case of ‘denaturalisation’ of the proposal is foreseen (while some were present in previous years).

36 As is well-known (among the first to see this, see O. Costa et al., Codecision and ‘early agreements’: An improvement or a subversion of the legislative procedure? (Notre Europe, no. 82 2011) especially. p. 15 ff), these agreements are becoming the norm in the ordinary legislative procedure. European Parliament, Codecision and Conciliation (2014), available at <www.europarl.europa.eu/code/information/guide_en.pdf>: in the previous legislative term (the seventh), p. 34: ‘93% of all adopted codecision files were ‘early agreements’ ‘compared to 54% and 82% during the 5th and 6th legislative terms, respectively’. Correspondingly, the number of files adopted at the third reading stage dropped ‘from 88 to 23 files between the 5th and the 6th legislative terms, while during the 7th legislative term only 9 files went to conciliation’. For data relating to the first half of the current legislative term see Tajani et al., supra n. 3, p. 10 (with a further increase of codecision files adopted either at first or early second reading: 97% overall, with a significantly higher proportion of files concluded at the early second reading stage, i.e. 22% of all adopted files). On trilogues see also supra, n. 25.

37 On different interpretations of this provision, and on its origins, see Chamon, M., ‘Upholding the ‘Community method’: Limits to the Commission’s Power to Withdraw Legislative Proposals. Council v Commission (C-409/13)’, 40(6) European Law Review (2015) p. 895 at p. 901 ff.

38 Cf Roederer-Rynning and Greenwood (2015), supra n. 25, p. 1148.

39 Cf Opinion of the AG, no. 57.

40 In the sense that the question of confidence exists in every parliamentary system, except Norway, Tsebelis, G., Veto Players. How Political Institutions Work (Princeton University Press 2002) p. 100 . See also Huber, J.D., ‘The Vote of Confidence in Parliamentary Democracies’, 90 The American Political Science Review (1996) p. 269 at p. 271 ff, who instead includes Norway and considers Iceland to be the main exception.

41 See Huber, supra n. 40, p. 271, noting that ‘in Belgium, Canada, Denmark, Germany, Great Britain, Ireland, New Zealand, Norway, Portugal, and the United Kingdom, there are no formal institutional constraints on the prime minister (although consultation with the cabinet is generally presumed)’. On the contrary, ‘in Finland, Italy, Spain, and Sweden, the prime minister must receive the formal support of the cabinet before a confidence vote procedure can be invoked’.

42 This is the case in Italy, where the question of confidence, although not explicitly addressed by the Constitution, is frequently posed by the government on ‘maxi-amendments’ in order to gather its parliamentary majority and achieve, through an open ballot, the approval of legislative bills: see Olivetti, M., La questione di fiducia nel sistema parlamentare italiano [The Question of Confidence in the Italian Parliamentary System] (Giuffrè 1996) and Gianniti, L. and Lupo, N., Corso di diritto parlamentare [The Course of Parliamentary Law] (Il mulino 2013) p. 215 ff.

43 The classic example is the French Fifth Republic, in which the government is entitled – directly by the Constitution – to use a series of procedural instruments to force and direct a parliamentary vote: see Avril, P. et al., Droit parlementaire (Lextenso 2014) p. 229 .

Also, in the Netherlands a Minister can declare an amendment ‘unacceptable’, thereby threatening the withdrawal of the bill or his own demission (or even the resignation of the entire government): see Bovend’Eert, P.P.T. and Kummeling, H.R.B.M., Het Nederlandse Parlement [The Dutch Parliament] (Wolters Kluwer 2017) p. 242 .

44 In political science terminology, these are all instruments by which the Executive controls the legislative agenda: cf Pollack, M.A., The Engines of European Integration. Delegation, Agency, and Agenda Setting in the EU (Oxford University Press 2003). For a wide and comparative analysis see Döring, H., Parliaments and Majority Rule in Western Europe (Mannheim Centre for European Social Research 1995).

45 The traditional principle is clearly stated in Pierre, E., Traité de droit politique, électoral et parlementaire 2 (Imprimeurs de la Chambre des Deputès 1902) p. 77 ff. See, for instance, for the Italian legal order, in which the government is allowed to withdraw its own legislative proposals (until they are approved by one of the two Houses): Cicconetti, S.M., ‘Il potere di ritiro nel procedimento di formazione delle leggi’ [The Power of Withdrawal in the Law-making Process], 15(2) Rivista trimestrale di diritto pubblico (1965) p. 381 (remarking that this power has been used mainly for technical reasons). For a general recognition of the power of withdrawal for any bill until both houses have adopted it, see Art. 86 of the Dutch Constitution.

46 Pizzorusso, A., ‘The Law-Making Process as a Juridical and Political Activity’, in A. Pizzorusso (ed.), Law in the Making. A Comparative Survey (Springer-Verlag 1988) p. 152 .

47 For a list, see Ponzano, supra n. 2, p. 39, who cites the only six cases between 1977 and 1994 in which the Commission exercised this power for political reasons. On the consensual nature of the ‘overwhelming majority of cases’ of withdrawals see Oliver, P. and Matenczuk, B., ‘The Commission’, in R. Schütze and T. Tridimas (eds.), Oxford Principles of European Union Law. Volume I: The European Union Legal Order (Oxford University Press 2018) p. 549 at p. 572.

48 Normally, this kind of withdrawal is the final step of a procedure that uses an online platform: <ec.europa.eu/info/law/law-making-process/overview-law-making-process/evaluating-and-improving-existing-laws/reducing-burdens-and-simplifying-law/refit-platform/refit-platform-work-progress_en>, visited 18 April 2018. For a recent assessment of this program, see the Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions, Completing the Better Regulation Agenda: Better solutions for better results (COM(2017) 651 final, 24 October 2017).

49 Cf Arts. 6 and 7 of Protocol No. 2 (on the application of the principles of subsidiarity and proportionality).

50 See, respectively, Fabbrini, F. and Granat, K., ‘“Yellow card, but no foul”: The role of the national parliaments under the Subsidiarity Protocol and the Commission proposal for an EU regulation on the right to strike’, 50 Common Market Law Review (2013) p. 115 ; , D. Fromage, , ‘The Second Yellow Card on the EPPO Proposal: An Encouraging Development for Member State Parliaments?’, 35(1) Yearbook of European Law (2016) p. 5 ; Jančić, D., ‘EU Law’s Grand Scheme on National Parliaments: The Third Yellow Card on Posted Workers and the Way Forward’, in D. Jančić (ed.), National Parliaments after the Lisbon Treaty and the Euro Crisis. Resilience or Resignation? (Oxford University Press 2017) p. 299 .

51 In case of the ‘orange card’, moreover, if the Commission (or the other initiator) decides to maintain the proposal it has to justify why in its view it complies with the principle of subsidiarity. See Art. 7, paras. 2 and 3, of Protocol No. 2.

52 Cf the Opinion of the AG, no. 31, according to whom ‘Article 7 of the protocol on subsidiarity demonstrates that the authors of the FEU Treaty intended the Commission’s right of withdrawal to be a general right’.

53 On the origins of the political dialogue, by which national parliaments can raise any kind of observation, including on the substance of the proposal, see Jančić, D., ‘The Barroso Initiative: Window Dressing or Democracy Boost?’, 8 Utrecht Law Review (2012) p. 78 . On its more recent development, cf Jančić, D., ‘The Game of Cards: National Parliaments in the EU and the Future of the Early Warning Mechanism and the Political Dialogue’, 52(4) Common Market Law Review (2015) p. 939 .

54 In the sense that the principle of subsidiarity is ‘politically complex and legally uncertain’ see de Búrca, G., ‘The principle of subsidiarity and the Court of Justice as an institutional actor’, 36 Journal of Common Market Studies (1998) p. 217 . Evans, M and Zimmermans, A, ‘Editors’ Conclusions: Future Directions for Subsidiarity’, in M. Evans and A. Zimmermans (eds.), Global Perspectives on Subsidiarity (Springer 2014) p. 221 ff at p. 223 talk of a principle that is ‘somewhat of a chameleon’.

55 See Goldoni, M. and Jonsson Cornell, A., ‘The Trajectory of the Early Warning System’, in A. Jonsson Cornell and M. Goldoni (eds.), National and Regional Parliaments in the EU-Legislative Procedure Post-Lisbon (Hart 2017) p. 335 ; I. Cooper, ‘Is the Early Warning Mechanism a Legal or a Political Procedure? Three Questions and a Typology’, ivi, 18-49 (remarking that the Early Warning System has been studied by both law and political science scholars).

56 See, although with different approaches, Kiiver, P., The Early Warning System for the Principle of Subsidiarity: Constitutional Theory and Empirical Reality (Routledge 2012) (referring to national parliaments as a sort of ‘Council of State’); and Fabbrini and Granat, supra n. 50, p. 115 (criticising the broad interpretation of the principle of subsidiarity adopted by national parliaments in the first yellow card).

57 See Goldoni, M., ‘The Early Warning System and the Monti II Regulation: The Case for a Political Interpretation’, 10(1) EuConst (2014) p. 90 and Lupo, N., ‘National parliaments in the European integration process: re-aligning politics and policies’, in M. Cartabia et al. (eds.), Democracy and Subsidiarity in the EU (Il mulino 2013) p. 107 .

58 The case of the third yellow card seems particularly telling, as it was used almost exclusively by the Parliaments of eastern and central European Member States to oppose a directive on posted workers which was opposed by their government: see Jančić, supra n. 50, p. 299, and Fromage, D. and Kreilinger, V., ‘National parliaments’ third yellow card and the struggle over the revision of the posted workers directive’, 10(1) European Journal of Legal Studies (2017) p. 125 .

59 See the letter sent by the then Vice-President of the Commission, Šefčovič, to all national parliaments on 12 September 2012.

60 See COM(2016) 505 final, 20 July 2016. For additional considerations see N. Lupo, ‘Le molteplici funzioni dell’Early Warning System, alla luce del terzo “cartellino giallo” sui lavoratori distaccati’ [The Many Functions of the Early Warning System, at the Light of the Third ‘Yellow Card’, on Posted Workers], in Dialoghi con Ugo Villani (Cacucci 2017) p. 583.

61 For some data, derived from the European Council’s conclusions, which often invite or welcome or look forward to an initiative by the Commission, see O. Höing and W. Wessels, ‘The European Commission’s Position in the post-Lisbon Institutional Balance. Secretariat or Partner to the European Council?’, in Chang and Monar, supra n. 5, p. 123 at p. 134.

62 Cf A. Vitorino, ‘Preface’ to Ponzano, supra n. 2.

63 Cf Rasmussen, A., ‘The Role of the European Commission in Co-decision. A strategic facilitator operating in a situation of structural disadvantage’, European Integration online Papers (EIoP), Vol. 7(2003), 10 , available at <eiop.or.at/eiop/texte/2003-010a.htm>, p. 9, visited 18 April 2018 (obviously referring to the codecision procedure). On more recent developments, along the same line, cf Burns, C. et al., ‘Legislative codecision and its impact on the political system of the European Union’, 20(7) Journal of European Public Policy (2013) p. 941 .

64 V.A. Schmidt, ‘The political dynamics of EU Governance in Response to Crisis: Toward More or Less Legitimacy?’, draft paper prepared for the Berlin Workshop (6-7 October 2017): ‘Beyond representative democracy? Design and legitimacy of non-majoritarian institutions’, p. 19, available at <www.democratic-anxieties.eu/berlin>, visited 18 April 2018.

65 A hint in this direction is offered by K. Lenaerts, ‘L’Évolution du Cadre Juridique-Institutionnel de l’Union Européenne’, in Tizzano supra n. 20, p. 5 at p. 9 ff, underlying the importance that the principle of institutional balance – as opposed to the principle of separation of powers – has in the EU institutional system, both at the stage of legislative initiative and, after the judgment of the Court of Justice, also at the time when the power to withdraw may be exercised.

* Full Professor of Public Law and Director of the Centre for Parliamentary Studies LUISS University, Rome. The author would like to thank the reviewers as well as Robert Bray for their comments on previous versions of this article.