Introduction
Since the entry into force of the Lisbon Treaty, EU soft law has become more prominent in the EU system. EU institutions are increasingly adopting soft law measures,Footnote 1 which may take the form, among others, of recommendations, notices, action plans, guidelines and opinions.Footnote 2 As observed by Stefan, the Covid-19 crisis has prompted the EU to adopt a plethora of EU soft law acts.Footnote 3 The growing issuance of these measures may find a justification in the light of the following characteristics of these acts. First, the adoption of EU soft law is less burdensome than the issuance of other instruments. Most EU soft law acts are not subject to strict procedural requirements and no extensive public debate is required for their adoption. Second, EU soft law may be issued virtually in any EU policy field, and even in areas where the EU does not have regulatory competences (yet).Footnote 4 The absence of features typical of EU legislative and non-legislative acts, such as procedural safeguards and limitations in terms of scope, signals the sui generis nature of EU soft law acts.
Not surprisingly, in recent years litigation over EU soft law measures has increased before the EUFootnote 5 and nationalFootnote 6 courts. In this context, a peculiar situation concerning the review of EU soft law before EU Courts may be observed. Recent EU case law has – almost systematically – excluded judicial review of these measures through the action for annulment (Article 263 TFEU). In a number of cases, the EU Courts reached the conclusion that the challenged EU soft law acts did not possess ‘legally binding effect’. As a consequence, actions for annulment regarding EU soft law acts have been dismissed. Remarkably, the production of ‘legally binding effects’ by EU soft law has been assessed against elements such as the form of the document or the intention of the authors, rather than the substance and implications of the challenged EU soft law instruments.
The preliminary ruling procedure has only partially contributed towards ensuring effective check and balances in relation to these acts. Through the preliminary ruling requests, the Court of Justice has identified certain legal effects deriving from these measures and mainly concerning the issuing authority. At the same time, the EU judicature has overlooked the impact of these acts on individuals, thus appearing reluctant to identify legal effects of EU soft law on third parties. In turn, individuals could not obtain clarifications as to the implications of these measures on their positions, while having limited access to judicial review of these acts.
In the light of this overview, it may be argued that EU soft law is partially escaping the ‘EU constitutional order’. In fact, EU soft law enables EU institutions to exert ‘soft regulatory powers’, currently not fully subject to the rules and principles of the EU, due to limited judicial scrutiny by the EU Courts.Footnote 7 However, so long as EU soft law measures are adopted, the need to ensure effective judicial review – namely, that the sui generis regulatory powers expressed through them comply with the EU constitutional framework – emerges powerfully.Footnote 8 The centrality of effective judicial review as an EU general principle is confirmed by its connection to the rule of law, one of the EU founding values protected under Article 2 TEU.Footnote 9
This paper argues that the Court of Justice of the European Union should open the gates of Article 263 TFEU to ensure effective judicial review of EU soft law measures. For this purpose, the paper advocates the adoption of a liberal-constitutional reading of the Treaties, and in particular, of Articles 263 and 288 TFEU with regards to the scrutiny over EU soft law. Liberal-constitutionalism refers to a constitutional theory that assumes the expansive nature of constitutions. The main objectives of this framework are the creation of check and balances among the traditional ‘state’ powers (i.e. legislative, executive and judiciary)Footnote 10 and the judicial protection of individual liberties and rights.
Under a liberal-constitutional reading, Articles 263 and 288 TFEU could be interpreted to allow direct judicial review of EU soft law while relaxing the restrictive requirement of ‘legally binding effects’. By subjecting EU soft law to effective judicial review via the action for annulment, the Court of Justice would safeguard effective judicial review and institutional balance in a coherent EU legal order. The direct review of EU soft law would offer the opportunity for the Court of Justice to mould the ‘limits’ of the EU constitutional order, both from a vertical perspective (EU–member states) as well as horizontal (among EU institutions). In this way, the sui generis regulatory powers of soft law would be put in check under the EU constitutional framework, and individuals, as well as member states and EU institutions, would be afforded more effective judicial review of these acts.
The paper is structured as follows. First, it provides an overview of liberal constitutionalism theory. Second, the paper demonstrates that the EU Courts’ case law is imbued with liberal-constitutional readings of the Treaties. The EU liberal-constitutional jurisprudence has focused on the centrality of judicial review and effective judicial protection of individual rights. Third, the paper discusses the loopholes in the current framework of review of EU soft law from the angle of the EU general principle of effective judicial review. Finally, the paper illustrates that possible drawbacks of extending the action for annulment to EU soft law fail to consider the centrality of effective judicial review in the EU, and do not constitute material risks for the efficiency of the EU institutions and the legitimacy of the EU judiciary.
What the paper does not wish to do is to argue for an unlimited judicial review for EU soft law acts. More modestly, it seeks to highlight some issues of the current review mechanisms applied to EU soft law and to offer the theoretical underpinnings to ensure direct review of EU soft law.
Liberal constitutionalism: an overview
Liberal constitutionalism is a constitutional theory characterised by three elements: first, the idea that the constitution is not statically sculpted in the wording of the law, but should be subject to the interpretation of the judiciary; second, the objective to rebalance the disequilibria existing between individuals and other state powers; third, powerful judicial control systems, with courts being the less intrusive power to draw the boundaries between the protection of individual rights and the scope of administrative action. The central objective of liberal constitutionalism is to limit ‘the inconvenience of [a]bsolute power’.Footnote 11
According to Loughlin,Footnote 12 liberal constitutionalism finds its origin in The Federalist no 78, where Hamilton suggested that the US Supreme Court should be entitled to have the power to annul all acts that are contrary to the tenor of the US constitution. This constitutional paradigm promotes the authority of an independent group to interpret and enforce the terms of the text of the constitutional document. According to Hamilton’s views, the limits of the constitutional text are flexible and subject to judicial interpretation; the Hamiltonian standpoint leads to a position in which the Constitution is what the judges say it is. Such a conception was enshrined in the pivotal US Supreme Court’s judgment Marbury v Madison,Footnote 13 establishing the US Supreme Court’s power of judicial review to limit the legislative and executive branches of the government.
The liberal-constitutional paradigm has been studied and substantively developed by Hayek.Footnote 14 This author speaksFootnote 15 of a ‘liberal constitutionalism’, the chief aim of which is ‘to provide institutional safeguards of individual freedom’ and ‘to secure individual liberty by constitutions’. In the reconstruction of Hayek, individual liberty (typical of liberal systems) and democracy could be reconciled when ‘all authority is restrained by long-run principles which the opinion of the people approved’. Courts are the most appropriate fora to set the limits of state authority and to say what the law is.Footnote 16 The advantage of liberal constitutional systems, in Hayek’s theory, is that they provide a notion of the law as a limit to power. The law is not a static entity, but rather malleable: judicial interpretation seeks to achieve equilibria between the individual and the central power under the oversight of the judiciary.Footnote 17 Ultimately, liberal constitutionalism ensures the flexibility of constitutional orders with the view to applying supreme principles of constitutional law to acts and conducts adopted within a system.Footnote 18
Therefore, there is overall consensus in literature that judges and, in particular, constitutional courts, are the key actors in liberal constitutionalism: first, they identify principles and values restraining the reach of public powers; and second, they apply the constitution to sui generis regulatory actions not envisaged under the basic norms of the legal order of reference. A crucial element contributing to liberal constitutionalism is the enforceability of fundamental rights. Being amongst the supreme rules of a legal order, fundamental rightsFootnote 19 may bind all actors of a given legal system, be they institutions, individuals or companiesFootnote 20. In this sense, fundamental rights may be used as a sword when striking down invalid legislation or executive actions, and as a shield when protecting or granting individual rights.
As a matter of fact, supranational bodies increasingly exercise public powers in parallel to national authorities.Footnote 21 Transnational arrangements cover topics such as financial regulation, competition, energy and trade policy, environmental protection, crime and security, and so on. At national level, the emergence of authorities with regulatory powers has entailed the adoption of new measures and acts whose consequences are often uncertain.Footnote 22 These developments, such as the emergence of quasi-governmental bodies whose competences are often not captured by existing accountability mechanisms, undermine the claims of modern constitutions to be comprehensive in their reach.Footnote 23 One response to this situation has been to loosen the ‘anchorage’ of constitutional norms, including fundamental rights, for the purpose of extending their application.Footnote 24 In this context, liberal-constitutional interpretations highlight a detachment of constitutions both from the idea that they are formalistic documents as well as from the ‘traditional’ state structure.
Therefore, liberal-constitutionalism may lead to interpretations of the law going beyond its wording. Such readings of the law have been applied with different degrees of intensity also beyond the United StatesFootnote 25 and on a wide range of matters. In adopting liberal-constitutional interpretations of their respective constitutions, courts have ensured checks and balances, as well as protection of individual rights, in their respective legal orders. It may be argued that liberal-constitutional interpretations aim to enhance coherence in the constitutional order and also respect of the supreme legal principles when it comes to sui generis governmental actions. As a consequence, liberal-constitutionalism does not merely entail judicial empowerment, but also effective judicial protection of individual positions vis-à-vis new regulatory powers. Overall, liberal constitutionalism signals the prevalence of the substance over the procedure in legal orders, under moral conceptions of fairness and justice.
Liberal constitutionalism has also shaped international legal orders.Footnote 26 A prominent example of a liberal-constitutional system where flexible readings of the basic laws have significantly occurred with the objective of ensuring checks and balances is, interestingly, the EU.Footnote 27
Liberal constitutionalism in the case law of the Court of Justice of the European Union
The constitutional nature of the EU encounters the support of both the Court of Justice’s case lawFootnote 28 and scholarship.Footnote 29 Schutze,Footnote 30 especially, has discussed how the EU constitutional order incorporates various aspects of constitutionalism, in particular, liberal-constitutionalism. The Court of Justice has been a central actor in ensuring the progressive constitutionalisation of the EU as a whole.Footnote 31 A peculiar feature of EU constitutionalisation is its bi-directional impact: it operates vis-à-vis the national legal orders, and within the EU itself. In the first dimension, constitutionalisation sought to legitimise the EU as an emerging power, limiting national sovereignty. It enhanced the constitutional ‘credibility’ of the EU as an atypical constitutional legal order. In the second form, constitutionalisation constrains the action of EU institutions within the EU legal order, in the respect of EU general principlesFootnote 32 and values.Footnote 33 All in all, the EU is the epitome of liberal constitutionalism.Footnote 34
The most evident liberal-constitutional element in the EU is in regard to the development of EU general principles.Footnote 35 Under the umbrella of general principles, proportionality, effectiveness, equivalence, non-discrimination and effective judicial protection have made their entrance in the pantheon of EU primary sources. The jurisprudential development of general principles of EU law led, in particular, to a double constitutionalisation: the EU legal order was expanded through the injection of moral-based values given by the general principles; and, further, actions both from the EUFootnote 36 and national authoritiesFootnote 37 have been reviewed and struck down because of incompatibility with the EU general principles. The Court of Justice’s case law on general principles of EU law is an additional proof of the fluidity attached to the EU constitutional settings and the pivotal role exerted by that Court in bringing rationality and coherence in the EU order.
The acquisition of legally binding effects by the EU Charter of Fundamental Rights has not reduced the importance of general principles.Footnote 38 On the contrary, the scope of general principles of EU law has been amplified by the presence of the Charter.Footnote 39 What is more, following the path initiated with the development of general principles of EU law, the Court of Justice has given broad application to the Charter of Fundamental Rights by way of a loose interpretationFootnote 40 of Article 51 thereof, defining the scope of application of the Charter in relation to EU and national measures.Footnote 41 The Court of Justice was thus able to extend the compass of EU fundamental rights over national actions.Footnote 42
Another liberal-constitutional feature of the EU is the centrality of the principle of effective judicial review, being an expression of the rule of law. In Les Verts Footnote 43 the Court of Justice, for the first time, relied on the principle of effective judicial review to allow scrutiny by EU Courts over acts of the European Parliament that are not listed in the Treaties among the acts amenable to judicial review. Ensuring control over all EU institutions’ acts is, according to the Court of Justice, an expression of the rule of law in the EU.Footnote 44 The Chernobyl caseFootnote 45 further stressed that the judicial review carried out by EU Courts was instrumental in ensuring institutional balance and, thus, effective separation of powers in the EU.Footnote 46 As a result of that case, the Parliament was deemed entitled to bring an action for annulment, although it was not listed among the entities authorised to bring such an action.
It is in the well-known UPA caseFootnote 47 that the Court of Justice made explicit the link between effective judicial review and the rule of law. In particular, it affirmed that ‘The European Community is a community based on the rule of law Footnote 48 in which its institutions are subject to judicial review of the compatibility of their acts with the Treaty and with the general principles of law which include fundamental rights’. The symbiosis between effective judicial review and the rule of law has been confirmed in recent EU jurisprudence. For instance, Juízes Portugueses Footnote 49 has indicated that the principle of effective judicial review is part of the essence of the rule of law, one of the founding values of the EU protected under Article 2 TEU.
Finally, a further indicator of liberal-constitutional elements in the EU order is the principle developed in the ERTA case,Footnote 50 according to which acts issued by EU institutions in the exercising of implied powersFootnote 51 that have legal effects may be subject to judicial review, even though these measures may not be listed as reviewable under the Treaties. This case law offers an additional example of the liberal-constitutional interpretations of EU Treaties and the need to ensure effective checks and balances also in the attainment of the EU’s objectives.
Overall, the considered jurisprudence extended the reach of the EU constitutional order and ensured the rationality and coherence of the EU legal order via flexible interpretations of the Treaties and judicial review: this is an intrinsically liberal-constitutional case law. The same approach, however, does not seem to apply in the EU case law on the judicial review of EU soft law.
Judicial review of EU soft law: recent jurisprudential developments and the need for effective direct judicial review
As mentioned above, EU soft law is the expression of sui generis regulatory powers. In this sense, EU soft law is one of the facets of the EU’s atypical action. The increasing issuance of EU soft law was already highlighted by Stefan in 2014, and this trend is continuing.Footnote 52 Statistics show that between 2019 and 2020 EU institutions have adopted 138 soft law instruments in the form of opinions and recommendations.Footnote 53 More recently, with the raging of the Covid-19 crisis, EU institutions have made extensive recourse to EU soft law. In particular, the Commission has issued a communication on a coordinated economic response to the Covid-19 outbreak,Footnote 54 dealing with the immediate response to the crisis, followed by two roadmaps by the Council and the CommissionFootnote 55 on strategies and measures to end the lockdowns.
ScholarsFootnote 56 have observed that EU soft law may induce certain individual behaviours and even modify normative reality.Footnote 57 Prime examples are the Press Release No 144/16 concerning the EU-Turkey agreement on migrants, or the Euro Group Statements issued in the context of the bail-in measures adopted in Cyprus and concerning certain banks. These instruments have been adopted as part of the implementation of EU actions at international and member state level and have thus impacted third parties.Footnote 58
Interestingly, the latest developments in EU case law indicate that these measures are subject to an ‘incomplete judicial scrutiny’ by the EU judicature. While the European Court of Justice has interpreted EU soft law through the preliminary procedure, the EU judicature has been reluctant to directly review EU soft law through actions for annulment.
After discussing the admissibility criterion of ‘legal effects’ under Article 263 TFEU, the following section will discuss, first, the latest decisions on the reviewability of EU soft law, both via actions for annulment and the preliminary ruling procedure, and second, the drawbacks of the current approach as to judicial scrutiny of these measures by EU Courts.
Judicial review in the EU: the notion of ‘legal effects’
Currently governed by Article 263 TFEU, judicial review of EU acts is initiated before the General Court at first instance and, on appeal, before the Court of Justice.Footnote 59 A central element for the assessment of whether an act could be directly scrutinised by the EU judicature is linked to the production of legal effects towards third parties.Footnote 60 In this regard, the Court of Justice of the EU has adopted a ‘substance over form’ approach, whereby the substance of an act should determine whether it produces legal effects. This case law is a further example of liberal-constitutionalist case law in the EU.Footnote 61
The ERTA judgmentFootnote 62 illustrates these points. In that case, the Court of Justice held that ‘an action for annulment must […] be available in the case of all measures adopted by the institutions, whatever their nature or form, which are intended to have legal effects’.Footnote 63 The Court found that the proceedings of the Council for the adoption of the ERTA agreement aimed to ‘lay down a course of action binding on both the institutions and the member states’. The Council’s submission – arguing that the member states’ coordination in negotiating the ERTA agreement was only ‘voluntary’, and not compulsory – could not affect the production of legal effects of the Council’s proceedings. As a result, the latter produced legal effects and could be reviewed by the EU Court.
Similarly, in Commission v Council,Footnote 64 the Court of Justice found the production of legal effects by a decision of the Council, providing the member states with voting powers at the UN. The fact that such a vote had implications on the relationships between the Community, the member states and the international order was liable to attribute legal effects to that act. Once again, the submission of the Council, which argued that the decision had only a procedural nature and was not binding, did not affect the presence of legal effects.
The irrelevance of the form or intention of the author of an act in determining the production of legal effects was confirmed in France v Commission.Footnote 65 The Court of Justice held that:
an action for annulment is available in the case of all measures adopted by the institutions, whatever their nature or form, which are intended to have legal effects. This applies to a Commission communication which sets out to specify the manner of application of Article 5(2) of Directive 80/723 on the transparency of financial relations between member states and public undertakings, which was published in the C Series of the Official Journal and was notified to each member state.Footnote 66
By detailing the way in which a directive provision had to be applied by the member states, the Commission’s communication was deemed to produce legal effects.
In IBM,Footnote 67 it was further specified that:
any measure the legal effects of which are binding on, and capable of affecting the interests of, the applicant by bringing about a distinct change in his legal position is an act or decision which may be the subject of an action under Article 173 for a declaration that it is void. However, the form in which such acts or decisions are cast is, in principle, immaterial as regards the question whether they are open to challenge under that article.
In the same case, the Court of Justice considered ‘the purpose’ of the EU act at stake (i.e. a statement of objection by the European Commission in the context of competition proceedings), to determine the legal effects and nature of the act. Since the act in question could not have a direct impact on the position of third parties, it could not be considered to produce legal effects.
Remarkably, the approach of the Court of Justice of the European Union as to the admissibility of direct actions concerning EU soft law seems to have (at least in part) departed from this well-established case law.
Direct judicial review of EU soft law: recent (restrictive) developments
When considering the admissibility of actions for annulment concerning EU soft law, in recent cases the EU Courts have set high thresholds for the production of legal effects. In so doing, they have especially focused on the intentionsFootnote 68 and the powersFootnote 69 of the authors, as well as the formFootnote 70 of the EU soft law acts.
For instance, a statement issued by the member state’s heads of government that are reuniting in the European Council is not an EU act producing legal effects. This is what has been established in NF and Others v European Council,Footnote 71 a case concerning the judicial review of the EU-Turkey statement of 18 March 2016 regulating migrants’ relocation between the EU and Turkey. Having recalled that EU acts are admissible for an action for annulment, subject to the condition that they produce legal effects vis-à-vis third parties,Footnote 72 the General Court considered whether the act could be attributed to the European Council. It observed that the choice of words utilised in the statement could not be liable to ‘alter the content or the legal nature of the procedure to which it relates, namely, an international summit’. The General Court further highlighted that, in the light of the wording used, the act had the objective to inform the public on the EU-Turkey migration policies. Moreover, the ‘inappropriate use of the expression “Members of the European Council” and the term “EU” in a press release’ cannot in any event bind the EU. In other words, the members states’ heads of government reuniting in the European Council did not have the intention to issue a document binding on the EU. The Court ultimately concluded that the statement could not be attributed to the European Council and, consequently, the action for annulment was rejected. The case was later confirmed on appeal.Footnote 73
The General Court has also considered that an act issued by an EU body not possessing decision-making powers under the Treaties could not produce legal effects. In Mallis,Footnote 74 the claimants had challenged the validity of the Eurogroup statement of 25 March 2013 concerning the macroeconomic adjustment programme for financial assistance to Cyprus. Following the issuance of this statement, the Cypriot Parliament adopted a series of bail-in measures regarding national banks.Footnote 75 When considering the admissibility of the claim, the General Court observed that, according to the established case law, ‘in order to determine whether an act or decision produces binding legal effects capable of affecting the interests of the applicant by bringing about a distinct change in his legal position, it is necessary to look to its substance’.Footnote 76 However, when carrying this assessment, the Court first focused on the competences of the Eurogroup – the author of the act – and found that this latter entity did not have decision-making powers pursuant to the Treaties. The Court then analysed the content of the document. Although it included ‘categorical statements’, the act of the Eurogroup could not be deemed as producing binding legal effects in the light of the ‘context’ of its adoption; namely, the lack of competences and powers by the Eurogroup set out in the Treaties deprived the act of any legal effects. Once again, the content of the act and its impact on third parties played a limited role in determining its effects. The Court of Justice dismissed the appeal lodged against the decision of the General Court,Footnote 77 and concurred with the view that the statement in question did not create legal obligations.Footnote 78
The reluctance of the EU Courts to engage in the direct review of EU soft law further emerges in Czech Republic v Commission.Footnote 79 The Czech Republic had challenged a letter of the Commission rejecting the request of that State not to make available to the EU budget some resources. The General Court recalled that the ‘production of binding effects’ is to be evaluated in the light of the substance of the act.Footnote 80 Elements to be assessed are the content of the act, interpreted in the light of the context and powers of the institution that adopted it.Footnote 81 Focusing on the powers of the Commission, the letter of the Commission was considered as deprived of legal effects, since that institution was not competent to adopt that act.Footnote 82 The action for annulment was therefore dismissed. This outcome is quite remarkable, for at least two reasons. First, the letter in question was phrased in prescriptive terms and so it was perceived to be from the Czech Republic, which brought an action of annulment against it. Therefore, the wording did not play any role in establishing the legally binding nature of the act in question. Second, the Court of Justice had previously established that the absence of competence for the EU to adopt an act does not exclude the reviewability of that measure.Footnote 83 An appeal against this decision is currently pending.Footnote 84
The Court of Justice has also excluded EU soft law from the remit of actions for annulment in Belgium v Commission.Footnote 85 In this appeal case, the Court of Justice confirmed a judgment delivered by the General Court in the context of an action for annulment against the Commission’s Recommendation 2014/478/EU, concerning consumer protection from online gambling. The Court of Justice found that the first instance court correctly evaluated the recommendation as non-legally binding for the purposes of Article 263 TFEU, and thus rightly rejected the action for annulment as inadmissible.
At first instance, the reasoning of the General CourtFootnote 86 was based on the wording of Article 288 TFEU, according to which recommendations and opinions of EU institutions do not have binding effects. As such, Recommendation 2014/478/EU was therefore excluded from the scope of application of Article 263 TFEU. Interestingly, the Court had applied the established case law according to which the production of legal effects of EU acts must be assessed in the light of ‘the substance of that act and […] on the basis of objective criteria, such as the content of that act, taking into account, as appropriate, the context in which it was adopted and the powers of the institution which adopted the act’.Footnote 87 The General Court took the view that the document was worded ‘mainly in non-mandatory terms’. It further noted that ‘the content of the contested recommendation […] shows that that act is not intended to have any binding legal effects Footnote 88 and that the Commission had no intention to confer such effects on it’.Footnote 89
On appeal, the Court of Justice distinguished the case at hand from the ERTA Footnote 90 and Les Verts Footnote 91 precedents. The Court held that the ERTA dictum, according to which admissibility for judicial review of a EU measure depends on whether it produces legal effects, could not be applied in the case at issue: the former case related to a Council deliberation recorded in the minutes of a meeting, and not a recommendation. Also, the principles established in Les Verts could not be relied upon,Footnote 92 in so far as that case concerned the lack of a provision under the Treaties regulating the reviewability of the EU parliament’s acts, and not, as in Belgium v Commission, the interpretation of a Treaty provision excluding binding effects of EU recommendations.Footnote 93 As a result, the recommendation could not be amenable to judicial review under Article 263 TFEU.
This judgment has three implications. First, it established that EU acts, including EU soft law, are reviewable if they produce ‘legally binding effects’ under a combined interpretation of Article 263 and 288 TFUE. Second, the production of legal effects is significantly influenced by the intention of the author of the act. Third, the form of an act also has influence in determining the production of legal effects, more than the wording and the context in which the act was adopted, as well as its implications.
The outcome in Belgium v Commission significantly narrows the liberal-constitutional case law of the EU Courts on judicial review and the production of legal effects of EU acts. The joint reading of Article 288 and 263 TFUE that was followed in that case is problematic from a number of perspectives. First of all, this interpretation of Articles 288 and 263 TFUE automatically leads to the exclusion of recommendations and opinions from the acts that can be reviewed under the action for annulment before the European Court of Justice. In Belgium v Commission, little if no attention was paid by the EU judges to the perception of the addressees of the recommendation and that of the implementing authorities at national level, nor to the impact of that act on individuals. As a result, the EU judicature falls within an interpretative loop, whereby the Court cannot find legal effects divergent from those already determined by the form of the measures.Footnote 94 This runs against the ERTA case lawFootnote 95 and the principle according to which the form of an act should not determine its legal effects.Footnote 96
Second, the exclusion of recommendations from the remit of judicial control, due to the absence of legally binding effects, is at odds with the production of binding effects by EU soft law on their author. It is indeed established that EU soft law measures may be binding on the EU institution that adopted them.Footnote 97 It is thus not clear, however, how the bindingness of an EU soft law act on the issuing authority may not also produce legal effects towards third parties. De facto, effects may also occur incidentally for individuals.Footnote 98
Third, and most importantly, individuals and member states affected by EU soft law acts are deprived of judicial protection. On the one hand, the admissibility requirements for the purposes of the action for annulment are interpreted strictly, with no possibility to access review of these acts. On the other hand, since EU soft law is not subject to transposition duties, it may not be possible to challenge any national measure implementing EU soft law.Footnote 99
As a result, it may be submitted that the EU Courts have adopted a ‘formalistic’ understanding of the notion of legally binding effects when it comes to EU soft law. In this context, they have focused on the intention and the powers of the issuing authority, or the form of the act, without considering the implications of EU soft law acts in relation to third parties. In turn, by excluding direct judicial review of EU soft law, this jurisprudence creates a gap in the EU judicial protection system in relation to such acts.
It should be further observed that, in the absence of direct judicial review, an action to establish the non-contractual liability of the EU could, nevertheless, be available for damages caused by selected EU soft law acts.Footnote 100 In Ledra, the Court of Justice found that the fact that the European Stability Mechanism is outside the EU legal order should not prevent the admissibility of a claim for damages created by the Commission and the European Central Bank in the process of adopting a memorandum of understanding.Footnote 101 Whether the dictum of this judgment extends to other EU soft law instruments is, however, unsettled.
Having considered the current limitations to bringing an action of annulment concerning EU soft law, we should now assess whether the indirect judicial review mechanisms available under the Treaties offer effective judicial protection in relation to EU soft law acts.
Indirect judicial review of EU soft law: gaps in the system of EU effective judicial protection?
Indirect judicial review of EU soft law may occur through the preliminary ruling procedureFootnote 102 or by pleading their illegality in the context of other actions brought before the EU Courts.Footnote 103 As observed by Türk, Article 277 TFEU allows indirect review of a wide range of acts, including notices of invitations to tender, staff rules or evaluation guides.Footnote 104 When examining the invalidity of EU law under Article 277 TFEU, the EU judicature does not require the production of legally binding effects.Footnote 105
Similarly, under the preliminary ruling procedure, the question of the ‘legally binding nature’ of the act does not bear any consequence on the admissibility of the matters referred from national Courts.Footnote 106 An example of the approach of the Court of Justice in this respect is the James Elliot case.Footnote 107 The EU Court ruled that the lack of binding effects of harmonised standards does not preclude the Court from ruling on their interpretation.
Likewise, in the first reference for a preliminary ruling submitted to the Court of Justice by the German Constitutional Court,Footnote 108 the question arose as to whether the European Central Bank had the power to adopt a programme for the purchase of government bonds on secondary markets (Outright Monetary Transactions programme), announced by way of a press release. Several governments had submitted that the question was inadmissible, since the European Central Bank’s press release was a preparatory act, and, in any event, did not possess legal effects.Footnote 109 In its judgment, the Court of Justice considered that the press release was not a preparatory act but, instead, a legal measure, and thus rejected the inadmissibility plea. What is more, the Court did not explicitly assess the production of legal effects by the press release. Nevertheless, the judgment delineated with great detail the two major legal implications of that document. In so doing, the Court of Justice did not consider the form of the document, but instead its substance and effects on the EU economy.
First, the Court interpreted the objectives of the Outright Monetary Transactions programme as outlined in the press release. That programme was deemed ‘capable of contributing to the stability of the euro area, which is a matter of economic policy’.Footnote 110 The press release and the policies outlined therein were therefore instrumental in the achievement of EU policies. Second, the Court explained that the programme required implementation to conclude ‘outright monetary transactions on secondary sovereign debt markets’. Such transactions took place through ‘one of the monetary policy instruments provided for by primary law’.Footnote 111 The meticulous analysis of the consequences of the European Central Bank’s press release in Gauweiler is in sharp contrast to the approach of the EU Courts in Belgium v Commission. As discussed, in this latter judgment, the Court of Justice has considered the production of legal effects of the Commission’s recommendation without taking into account the consequences entailed by that document.
Through preliminary ruling interpretations, the Court of Justice has identified some legal effects attached to soft law measures.Footnote 112 In Dansk Rørindustri A/S,Footnote 113 the Guidelines on the method of setting fines were found to bind the discretion of the EU Commission. In Expedia,Footnote 114 the de minimis notice of the EU Commission bound the discretion of this institution and offered guidance ‘to the Courts and authorities of the member states in their application of […] article [101 TFEU]’. Furthermore, in Grimaldi,Footnote 115 national courts were asked to take into account EU recommendations ‘in order to decide disputes submitted to them’. In a subsequent preliminary ruling,Footnote 116 however, the Court held that the Banking Sector communication was not binding upon the member states but was merely limiting the discretion of the Commission.
It may be concluded that the Court of Justice’s approach to EU soft law through preliminary rulings diverges from that followed in the action for annulment. On the one hand, in the context of indirect judicial review, the EU judicature has indicated that certain EU soft law acts may have legal effects, for instance, in the form of interpretative documentsFootnote 117 or as a limit to the discretion of the issuing authority.Footnote 118 On the other hand, the EU Courts have often excluded these instruments from direct judicial review due to the lack of ‘legally binding effects’.Footnote 119 Therefore, in principle, EU soft law may be found to have ‘legal effects’ or ‘legally binding effects’ (or none) depending on the action brought, and against whom the effects of EU soft law are tested. Whether the legal effects identified in the context of preliminary ruling questions are the same as ‘binding legal effects’, as required under Article 263 and 288 TFEU for the purposes of the action for annulment, is not settled. The case law suggests that legal effects and legally binding effects are two distinct categories.Footnote 120 Yet, questions may arise as to whether this distinction is sound and practical.
In the light of this discussion, it is submitted that the preliminary ruling procedure cannot ensure an effective judicial review of EU soft law measures for the following reasons. First, the preliminary ruling procedure is not subject to strict time limits and thus allows the consolidation of the legal effects of EU soft law. It is true that the Court of Justice has argued that the preliminary ruling should be available to individuals who would not have locus standi under Article 263 TFEU.Footnote 121 This procedure thus contributes to ensuring scrutiny over EU acts. However, the strict time limits of the action for annulment ensure that an act may be challenged before its extensive application, both at EU and national level. The action for annulment would better ensure the uniform effects of EU soft law in the member states.
Second, the preliminary rulings on EU soft law have so far offered a fragmented interpretation of the legal effects of EU soft law measures, and only partially settled the legal uncertainty surrounding the implications of these acts. The Court of Justice has found such measures as binding only the authorFootnote 122 but not third parties; in parallel, EU soft law may also offer guidance to courts and authorities.Footnote 123 In this regard, by finding that the EU soft law binds its authors, the Court of Justice seems to overlook the effects of such acts on individuals. The author should, in fact, respect the EU soft law measure during its interactions with individuals. As a result, third parties are (indirectly) subject to the consequences of EU soft law. The case law issued following preliminary rulings has therefore enhanced legal uncertainty and has not consistently taken into account how authorities and individuals have perceived or have been affected by EU soft law.Footnote 124 EU soft law indeed applies in both the EU and national legal order and may produce different effects depending on the perception of the authority in charge of implementing it.Footnote 125 Overall, the lack of clarity about the effects of EU soft law may have adverse effects on the protection of legitimate expectations and the uniform application of EU law.
Third, and as a consequence, the preliminary ruling procedure cannot effectively eliminate, amend or restore the consequences of EU soft law acts in the member states.Footnote 126 National authorities may perceive EU soft law acts as binding (or not) and thus apply them in different ways. For instance, whenever national authorities do consider themselves as bound by EU soft law, and such measures are later found not to have legal effects through a preliminary ruling decision of the Court of Justice, ensuring compliance at national level with the Court’s judgment may be challenging, and may go against the principles of legal certainty and legitimate expectations. In conclusion, it is submitted that the preliminary ruling cannot ensure the uniform application and effective judicial review of EU soft law.
Advocating for a liberal-constitutional stance as to the direct judicial review of EU soft law under Article 263 TFEU: ensuring institutional balance and effective judicial protection in relation to EU soft law
Although indirect judicial review through preliminary ruling requests has its virtues, the absence of direct judicial review over the sui generis regulatory powers expressed via EU soft law engenders disequilibria in the EU legal order. First, the absence of direct judicial review entails an imbalance in favour of the EU ‘administrative/executive’ power vis-à-vis the EU legislator. EU soft law may undermine the principle of institutional balance in the EU order, with the risk for executive agencies and institutions to take over competences left to the EU legislative power.Footnote 127 Second, the absence of direct scrutiny also upsets the relationship between individuals and the EU, based on the possibility of obtaining effective judicial protection in the fields covered by EU law.Footnote 128 Indeed, EU soft law may influence individuals’ behaviour and their legal position vis-à-vis national and EU institutions, but they might not be subject to direct judicial review.Footnote 129 Finally, EU soft law creates friction between administrative and judicial powers. EU soft law measures tend to be considered as non-binding by their authors, and this has led the EU Courts to exclude the production of legally binding effects under Article 263 and 288 TFEU. Overall, the exclusion of direct judicial review over EU soft law measures has limited the role of the Court of Justice in addressing the question of constitutional relevance concerning these acts.
To solve these drawbacks, the view of the author is that the formalistic understanding of the notion of ‘legally binding effects’, for the purposes of the action for annulment, should leave space for a liberal-constitutionalist reading of Articles 263 and 288 TFEU, so as to also allow direct scrutiny over EU soft law. Direct review of EU soft law would address the limitationsFootnote 130 of the preliminary ruling procedure discussed above: thanks to stricter time limits on the annulment action, the effects of EU soft law could be ‘tested’ before they consolidate in the member states; furthermore, scrutiny over the adoption of these measures could be better ensured, jointly with enhanced judicial protection.
The above views are not intended to hide the fact that a more relaxed reading of the Treaties on the admissibility requirements for the judicial review of EU soft law may also have adverse consequences. For instance, a possible litigation flood concerning EU soft law may arise, with a negative impact on the effectiveness of the procedures before the Court of Justice. The risk of a litigation flood is also an underlying reason for the European Court’s restrictive approach to individual’s locus standi under Article 263 TFEU.Footnote 131 However, there is extensive literature proving that this danger is not, in fact, a real one.Footnote 132 Furthermore, when observing the national systems where more relaxed requirements for judicial review exist, no significant litigation floods may be traced.Footnote 133
Another possible drawback is that the judicial review of EU soft law could hinder the activities of the EU and thus block prompt responses to crises. The massive adoption of EU soft law in relation to Covid-19 suggests that the EU soft law is issued by EU institutions to also tackle emergencies. In these cases, triggering legislative procedures could slow down the institutional action in the EU, and EU soft law measures are preferred by the EU institutions to adopt swift manoeuvres. Therefore, it could be argued that, as a principle, EU soft law should be shielded from judicial review, with a view to achieving the effectiveness of the EU action.
This argument inevitably fails to take into account that any administrative or quasi-administrative measure should, in all circumstances, be mindful of individual rights and general principles of the EU legal order. Ensuring effective judicial review of EU measures is, as demonstrated, a corollary of the rule of law in the EU. Therefore, the effectiveness of the EU order should not prevail over ensuring its founding values. The achievement of policy objectives, even in situations of emergency, should be subject to rule of law guarantees,Footnote 134 courts having a crucial role in ensuring respect of such safeguards. Without the exercise of judicial review of EU soft law, the EU Courts cannot ensure compliance with these guarantees. As Curtin put by, ‘judicial review, to a great extent, is a substitute for the lack of political representativeness, participation, transparency – or generally speaking – political accountability […]’.Footnote 135 This especially applies in relation to EU soft law measures, whose adoption may not be subject to democratic debate.
Finally, a further potential criticism is that a liberal-constitutional reading of Articles 263 and 288 TFEU to ensure judicial review of EU soft law may be a manifestation of judicial activism. In relation to this argument, two points should be raised. First, it should be observed that Articles 263 and 288 TFEU do not provide details on concepts, such as those of ‘binding effects’ and ‘EU act’, that are relevant to assess whether an action for annulment is admissible. Therefore, it is for the EU Courts to interpret these notions, in compliance with their duty to ensure observance of EU law.Footnote 136 Not engaging in the interpretation of these concepts may lead to a denial of justice. Second, and not surprisingly, the restrictive notion of ‘legally binding effects’ currently used by the EU Court does not result from the wording of the Treaties, being itself a judicial construct. It is true that Article 263 TFEU excludes recommendations and opinions from the list of reviewable acts. Yet, other EU soft law instruments, such as guidelines, statements and press releases, are not per se excluded from judicial review according to the wording of Articles 263 and 288 TFEU. In any event, it is for the EU judicature to identify EU acts amenable to scrutiny under Article 263 TFEU, and where a measure issued by the EU institutions has legal implications on third parties (i.e. legal effects) it should be considered as an EU reviewable act under Article 263 TFEU.Footnote 137 It follows that also this criticism should be dismissed, since the interpretation of notions such as ‘binding effects’ and ‘EU acts’ included in Articles 263 and 288 TFEU is left to the EU Courts. In interpreting these concepts, the EU judicature cannot be accused of judicial activism, and is entitled to adopt a liberal-constitutional approach in the absence of limitations to this effect under the Treaties.
To conclude, an enhanced judicial control by the EU Court of Justice over EU soft law via direct challenges contributes towards solving the uncertainty concerning the implications of EU soft law. Effective judicial review of EU soft law measures would bring the EU institutions’ powers, expressed via soft law tools, under the EU constitutional framework. Direct review of EU soft law would offer a key instrument for the Court of Justice to subject EU soft law to the rules and principles stemming from the Treaties.Footnote 138 What is more, extending a liberal-constitutional reading to Articles 263 and 288 TFEU would also be compatible with the established interpretative methodology of the European Court of Justice. Liberal-constitutional readings of the Treaties are numerous and well-settled in the EU case law. This case law is distinctive in so far as the European Court of Justice interpreted the EU constitutional order as flexible, thus as not strictly determined by the wording of the Treaties, with the consequent need to ensure judicial review over EU actions undertaken beyond the remit of the EU Treaties.
Concluding remarks
This paper had finite but ambitious objectives: first, to discuss the issues related to the limited judicial review of EU soft law by the EU Courts; and second, to advocate for a liberal-constitutional interpretation of Articles 263 and 288 TFEU as to the admissibility of actions for annulment concerning EU soft law measures.
For these purposes, the paper has first discussed how EU soft law and the sui generis regulatory powers expressed therein currently escape the EU constitutional order. The paper has then delineated the notion of liberal-constitutionalism as a theory that ensures effective checks and balances on public authority and protection of individual rights and liberties through judicial review. Interestingly, the EU jurisprudence is imbued with liberal-constitutional readings. Examples of this methodology in the EU case law abound (e.g. the case law on EU general principles and the centrality of effective judicial review in the EU). These decisions have led to the ‘expansion’ of the EU constitutional order and, in parallel, the scrutiny of the Court if Justice over EU (seldom atypical) acts. Interestingly, the liberal-constitutional case law of the EU also includes a ‘substance over form’ approach as to the determination of the legal effects of EU acts for the purposes of the action for annulment.
Subsequently, the paper has analysed the recent case law concerning judicial review of EU soft law via actions for annulment. It has highlighted the formalistic notion of ‘legally binding effects’ employed to assess the admissibility of such actions. In this regard, the EU Courts have focused on the intention of the authors (e.g. NF v European Council) and their powers (Czech Republic v Commission and Mallis), as well as the form of the act (e.g. Belgium v Commission), to consider the challenged EU soft law measures as deprived of legally binding effects and thus not amenable to judicial review. The paper has highlighted how this case law diverges from the established EU liberal-constitutional jurisprudence; it has further discussed the adverse impact of these recent decisions on effective judicial review of EU soft law.
The paper has then moved on to the assessment of whether the preliminary ruling procedure can offer effective judicial protection in relation to EU soft law. After illustrating that the production of legal effects by EU soft law is irrelevant for the admissibility of preliminary rulings concerning these acts, the paper has answered this question in the negative for a number of reasons. Among others, the paper has identified the absence of strict time limits under this procedure and the impossibility for preliminary ruling decisions to repair or erase the consolidated consequences of EU soft law acts in the member states. In addition, indirect challenges of these measures by national courts allow for legal uncertainty to persist among individuals, national authorities and EU institutions.
To improve judicial protection in relation to EU soft law, this paper suggests that the formalistic stance of the Court of Justice on direct review of EU soft law should be abandoned to follow a more liberal-constitutional oriented interpretation of Articles 263 and 288 TFEU, with the view of allowing direct scrutiny of these acts through the action of annulment. Ensuring direct judicial review of these measures would allow the EU judicature to scrutinise the activities of EU institutions enshrined in EU soft law; additionally, individuals affected by EU soft law measures could obtain judicial protection. The paper has demonstrated that a more liberal approach towards judicial review of EU soft law would not affect the correct functioning of the EU judiciary and the effectiveness of the action of the EU, and would not entail judicial activism.
Additionally, applying a liberal-constitutional methodology to ensure the review of EU soft law under Article 263 TFEU would be in line with the EU Courts’ settled case law. Considering the EU judicature’s constitutional practice, no legal reason for excluding effective judicial review of EU soft law via the action from annulment may be identified. The substantial adoption of EU soft law in the context of the Covid-19 crisis might require a rethinking of the current reviewability of EU soft law.
