Rethinking direct effect and its evolution: a proposal

Abstract

This article revisits the present and future of the direct effect principle and submits solutions for its appropriate understanding and enforcement. Once the topic has been put into context and it has been shown why direct effect is an evolving notion whose scope goes beyond the Van Gend & Loos judgement and the doctrine originating from it, the study presents five intertwined arguments. First, direct effect has two facets since, aside from a subjective-substitutive form of direct effect, there exists an objective-oppositive manifestation of direct effect, whereby a directly effective European Union (EU) provision entails the disapplication of national law, without either conferring immediately an individual right or replacing the domestic norm in governing the case at hand. Second, the article claims that the obligation to disapply is triggered always by primacy and direct effect, never by primacy alone. In fact, justifying disapplication on the basis only of the primacy principle and not on primacy and direct effect is likely to undermine the logic implied in the relationship between the EU legal order and domestic legal systems. Third, there are legitimate derogations from the obligation to immediately disapply a conflicting domestic provision with EU law endowed with direct effect in so far as they are admitted by the Court of Justice of the European Union (CJEU) on the condition that the replacement of a national provision by an EU norm may harm individual rights and/or question the national identity clause enshrined in Article 4(2) Treaty on European Union (TEU). Fourth, the test on clarity, precision and unconditionality for assessing whether an EU norm is directly effective, as it was conceived by the CJEU, is no longer a pillar in the conceptualisation and practice of direct effect. As a matter of fact, only unconditionality, in practice, proves to be the core element of direct effect. Moreover, direct effect and direct applicability are equivalent concepts since unconditionality, as the condicio sine qua non of direct effect, coincides with direct applicability. Fifth, a distinctive quality of direct effect, along with unconditionality, is the creation of an advantage resulting from the application of EU law and the subsequent disapplication of national law. This implies that an EU directly effective provision can never be only in malam partem (ie, detrimental for the individual) and thus disapplication, lacking the existence of an advantage for the individual, shall never come into play. The article concludes that the duty to refine the doctrine of direct effect must be performed, ultimately, by the CJEU. Indeed, only the CJEU can offer guidance to national authorities, since it has an interpretive monopoly on the ifs, whens and hows of direct effect. To this end, it is vital that the EU judges, by rejecting argumentative minimalism brought to its extreme, come to reassert their constitutional role and re-establish the common core of the EU system, beginning with the principles that created and shaped it.

1. Introduction

The tireless critics of the European Union (EU) have a natural inclination towards underlining the fragilities and incongruences of both the EU legal system and the process of integration. In doing so, either knowingly or unknowingly, they tend to overlook the peculiarities and points of strength of the European legal order, as well as the extraordinary legal innovations − amongst international organisations − that were brought forth by the founding of the European (Economic) Communities.

I find this narrative ¹ puzzling because the above-mentioned peculiarities, points of strength, and innovations constitute the axis around which the whole integration process was formed and continues to evolve. Indeed, they represent both its original core and the principal factors of its development.

The reference goes first and foremost to the judiciary-based principles governing the interplay between EU law and domestic legal orders: effectiveness, primacy, effective judicial protection, direct effect, consistent interpretation. These principles express their strength and voice their effet utile thanks to a distinctive tool of judicial cooperation between Member States and the EU, ie, the preliminary ruling procedure laid down in Article 267 TFEU.

My reasoning, which partially fits into the tradition of the ‘pure theory of law’, or in a nuanced form of it, ² while being imbued with normative and doctrinal intakes, delves into the substantive implications for the individuals stemming from the way the relationships between national legal orders and EU law are (or shall be) designed.

In this framework, I have chosen here to reflect on a unique characteristic of the EU, ie, the content, ratio, scope, and extension of a principle that has, over the years, to an extent greater than any other, shaped the law of the EU, moulding and transforming it, in synergy with other principles and theories. I refer to the doctrine of direct effect, which, since its conceptualisation by the Court of Justice (CJ) in the Van Gend & Loos ruling, has allowed the EU legal system to become ‘a new legal order’, ³ and, in its interaction with the primacy principle established in Costa, provided the EU with a ‘legal system of its own which […] became an integral part of the legal systems of the Member States’. ⁴ Indeed, direct effect enables natural and legal persons to directly invoke, and rely on, EU primary and secondary law in order to challenge discordant domestic laws before national administrations and jurisdictions. To this end, direct effect, while placing natural and legal persons at the centre of the Treaties, determines the autonomy and originality of EU law, as it is its driving force and raison d’être. Due to direct effect, the international law axiom, according to which individuals lack legal personality and are generally relegated to the peripheries with respect to the enforcement and exercise of rights, in fact, is overturned and radically questioned.

In Van Gend & Loos, the reasoning of the Court was inspired by ‘une certaine idée d’Europe’ rather than by strictly technical arguments on points of law. ⁵ The Court’s decision to interpret EU law in the sense of conferring, even implicitly, rights on individuals, as the counterpart of obligations imposed upon the Member States, was ultimately a political choice, since it expressed the purposes and values that, in the view of the Court, were to direct the process of European integration. ⁶

On the other hand, the fact that direct effect is the backbone of EU law does not necessarily mean that such doctrine, nowadays, is clearly legally framed by the Court of Justice of the EU (CJEU, comprising both the CJ and the General Court) in its case law. It is not, in my opinion, despite its pivotal status within the EU legal framework. It should be, in fact, since grasping if, when, and how direct effect arises affects the individuals’ entitlement to invoke EU law and put it into action.

In this connection, the aim of my article is to reconstruct and revisit the present and future of the direct effect doctrine and shed light on what I deem to be its most problematic features, in particular when it engages with the principle of primacy, as well as with the remedy of disapplication. At the heart of the contribution lies the relation between EU law and national law, while the application of international law within the EU legal order will not be discussed. ⁷ Furthermore, neither the so-called incidental effect of directives (and their lack of horizontal direct effect), nor the horizontality of the Charter of Fundamental Rights − albeit both partially addressed in this article − will be investigated in depth. ⁸

In section 2, I will set forth the problem and explain why direct effect is an evolving, many-sided, and polysemic notion that cannot be traced back solely to the Van Gend & Loos judgement and the doctrine originating from it. In section 3, I will demonstrate that direct effect does not always come along either with the immediate conferral of a right by an EU norm or with the capacity of the latter to govern the case through the replacement of a conflicting national provision. In section 4, I will investigate the interplay between direct effect, primacy, and disapplication, identify the (legitimate) derogations from the obligation to disapply, and demonstrate that justifying disapplication on the basis of primacy alone, not of primacy and direct effect, entails a blank proxy for EU institutions and is, ultimately, capable of undermining the logic governing the bond between EU law and domestic legal systems, as well as the cooperation between EU and national authorities. In section 5, I will focus on the limits of the test on precision, clarity, and unconditionality, which was conceived and is still (at random, in fact) employed by the CJEU. To this end, I will examine in depth the interplay between the concepts of direct effect and direct applicability, and the relevance of the unconditional character of an EU provision as a precondition for asserting its direct effect. In section 6, I will identify which are, today, the true conditions of direct effect. In this respect, I will argue that a distinctive premise of direct effect, along with unconditionality, is the creation of an advantage resulting from the application of EU law and the subsequent disapplication of national law. This implies that an EU directly effective provision can never be only in malam partem (ie, purely detrimental for the individual) and thus disapplication, lacking the existence of an advantage for the individual, shall never come into play. The essay ends, in section 7, with several considerations on the future of direct effect and its effet utile, as well as on the role of the CJEU in its reconstruction and enforcement.

2. Direct effect beyond Van Gend & Loos: a concept in flux

As pointed out by the CJEU, the Van Gend & Loos judgement shall be deemed as a ‘source of and a framework for the principles which have shaped the constitutional structure of the European Union’. ⁹ Understanding the extent up to which Article 12 of the Treaty establishing the European Economic Community (TEEC) contained a ‘clear and unconditional obligation’ and, thus, was endowed with ‘internal immediate efficacy’, ¹⁰ pushed the CJ to evaluate whether Community law belonged to international law and, thus, if the TEEC was to be considered as a ‘standard international treaty’. ¹¹ As anticipated in the previous section, the CJ, in clarifying that Article 12 TEEC ‘produces direct effect and creates individual rights which national courts must protect’, ¹² defined the Community as ‘a new legal order of international law for the benefit of which the states have limited their sovereign rights, albeit within limited fields, and the subjects of which comprise not only Member States but also their nationals’. ¹³ In this context, the Van Gend & Loos judgement cannot be regarded simply as a historic ruling of the CJ. In fact, it has given way to a doctrine which has guided, and continues to guide, the case law of the CJEU, by complementing and fostering other core principles of EU law.

In the literature, what largely brings together academics is the description of direct effect as an indispensable tool of integration with a knock-on effect for completing and refining the process of European integration. ¹⁴ For that matter, there is agreement on the past of direct effect. Regarding the present and the future, however, there is no consensus. As to the present, there are divergences regarding how the doctrine of direct effect is interpreted and applied by the CJEU. As to the future, there is no consensus on the role that the doctrine could play in the relationships between EU and domestic law: several scholars keep underlining its potential, while others stress its limits or even its outdatedness.

The existence of a variety of doctrinal approaches shows to what extent the concept of direct effect is ever-changing ¹⁵ and ‘diluted’. ¹⁶ Direct effect is a ‘chameleon concept’ ¹⁷ and a multi-faceted juridical category. ¹⁸ It is not ‘carved in stone’ ¹⁹ and has been subject, in the EU case law, to a progressive ‘diversification et gradation’ ²⁰ of its effects. In this sense, it is a principle with respect to which the CJEU case law is neither clear nor precise. ²¹ Hence, direct effect is an imprecise notion, difficult to frame in a legal sense, particularly in its scope, boundaries and objectives. In some cases, a risk arises of ‘charging’ with a certain meaning the CJEU’s choice to employ one word rather than another in defining a directly effective EU provision and, therefore, a risk of falling into what has been defined a ‘nominalist fallacy’. ²² It has been remarked that this uncertainty would transform direct effect from a legal cornerstone of EU law into a ‘virus’, ²³ capable of potentially ‘infecting’ any theoretical elaboration regarding the interaction between EU law and the legal systems of the Member States.

Furthermore, the doctrinal disagreements also ensue from the asymmetries existing between national legal traditions: the variety of interpretations given to the concept of ‘individual right’ is proof of this phenomenon. ²⁴ In short, direct effect, even though it is ‘a Community concept par excellence’, ²⁵ reveals different facets depending on the relevant national legal tradition (be it doctrinal or jurisprudential), which, inevitably, represents a starting point for the legal interpreter and practitioner.

In conclusion, direct effect is a naturally dynamic concept, as are the legal categories of jurisprudential origin, through which ever-developing social phenomena are legally framed. From this perspective, there is no reason why direct effect should necessarily be limited to that which resulted from the Van Gend & Loos judgement and doctrine. On the contrary, it has developed into a broader legal category than it was at the outset of the process of European legal integration. Such a change of perspective should be understood, voiced, and systematised by the CJEU, given that, as was stressed by one author in particular, the Van Gend & Loos doctrine ‘seems to grasp only a thin fragment of EU law enforcement issues’ and that it ‘no longer gives an accurate idea of the ways through which EU law penetrates Member States through its enforcement in national courts’. ²⁶ As a matter of fact, even though − from Van Gend & Loos onwards − several constitutive and recurring elements have consolidated over the decades, an originalist understanding of the Van Gend & Loos doctrine is misleading since it neglects the transformation that occurred in this domain and gave rise to a constant fulfilment of direct effect’s content and scope, as well as to its partial reshaping. ²⁷ In this connection, as anticipated in the Introduction, five matters warrant in-depth analysis: the role of rights within the realm of direct effect (section 3); the relation between direct effect, primacy and disapplication (section 4); the test employed by the CJEU for assessing the directly effective nature of EU norms (section 5); the identification of direct effect’s true core elements (section 6); and the reasons for its useful effect still today (section 7).

3. The objective-oppositive facet of direct effect

A. The decoupling between direct effect and the conferral of rights with substitutive effects in the EU case law

In the Van Gend & Loos ruling, the conferral of a right is a core element of direct effect, hence the subjective dimension of direct effect. As a corollary to such conferral − that was implicitly inferred by the CJEU − EU directly effective provisions, which are invoked by natural and legal persons and attribute to the latter actionable rights, are capable of replacing domestic norms, when it comes to governing a case under the scrutiny of a national court, thereby producing substitutive effects. Disapplication, in light of the Van Gend & Loos judgement and the doctrine deriving from it, therefore, presupposes the existence of a right created by EU law and entails the application of an EU norm instead of a domestic provision.

As will be clarified in this section, the above-mentioned conventional understanding of direct effect, while valid for the Van Gend & Loos ruling, cannot be generalised. After all, direct effect does not necessarily mean that rights (explicitly or implicitly) are directly attributed to on individuals, nor does it always entail, per se, substitutive effects, whereby a norm of EU law always replaces a provision of national law. Direct effect has a different, objective, facet, in so far as it also reveals the capacity of an EU law provision to serve as a parameter of legality of national law, with exclusionary effects, regardless of the creation of a right stemming directly from the EU legal order. ²⁸ It is the so-called invocabilité d’exclusion, ²⁹ ie, the objective-oppositive manifestation of direct effect. In other words, in a number of situations the individuals are not immediately the addressees of rights recognised under EU law and suited to replace national provisions. What the individuals claim, in practice, is the obligation enshrined in EU law. Thus, they invoke an EU norm not to exercise a right attached to it, but rather to obtain a review of legality of a national provision, an administrative measure, or a judicial decision. In doing this, individuals seek to vindicate an advantage/interest through disapplication. Such review may lead to the disapplication of domestic rules, administrative actions, or judgements/orders, should there be a conflict with EU law. As will be further discussed, the advantage entailed by disapplication shall be intended as an interest or a ‘privilege’, borrowing from Hohfeld’s conceptualisation of ‘subjective legal situations’. ³⁰

The complexity of the matter is increased due to the EU case law, which is far from straightforward. Moreover, the question of the boundaries between the different forms of direct effect (subjective, objective, substitutive, oppositive), which has been examined or discussed by a few Advocates General, ³¹ has not been explicitly dealt with at all by the CJEU to this day. There is, therefore, great uncertainty on the matter.

In this light, the CJEU’s case law, notably the cases regarding directives, will be interpreted, in this section, in the sense of favouring a broad interpretation of direct effect that is inclusive of objective-oppositive direct effect.

Against this background, the starting point in the process of objectification of direct effect is the Becker judgement, ³² regarding Article 13, part B, letter d), No. 1, of the Sixth Directive 77/388/EEC on the harmonisation of Member States’ law in relation to the turnover tax. ³³ In paragraph 25, the existence of two manifestations of direct effect (subjective and objective) ³⁴ is clear where it is stated that, wherever the provisions of a directive appear to be unconditional and sufficiently precise, those provisions may ‘be relied upon as against any national provision which is incompatible with the directive or in so far as the provisions define rights which individuals are able to assert against the State’. This formula has been reiterated multiple times in the case law of the CJEU. ³⁵

In the EU case law there are two types of cases from which the existence of a dichotomy between subjective-substitutive direct effect and objective-oppositive direct effect can be inferred: a) the national legal order comprises a derogation, ie, a provision that deviates from a rule enshrined in the domestic legal order, causing a breach of EU law; b) the national legal order comprises trade-related technical rules amending the domestic technical rules already in force, which were not notified by the State to the Commission, in contrast with EU obligations, and are thus unenforceable in judicial proceedings.

In both situations, the individual relies on EU law and seeks the disapplication of national provisions. There is neither the direct conferral of a right stemming from an EU provision nor the replacement of national law by EU law. Indeed, the case at hand is governed by domestic provisions which remain applicable since they are not affected by the remedy of disapplication. In this vein, as pointed out in the literature, EU law ‘immunises’ the individual from a provision or an action of the State ³⁶ and, in doing so, serves as a ‘shield’ rather than a ‘sword’. ³⁷

As examples of the first strand of rulings, two are worth mentioning. The first is Johnston, ³⁸ which regards the interpretation of Directive 76/207/EEC on equal treatment of men and women. ³⁹ In the judgement, the CJ first discussed the interpretation of national law in conformity with EU law. ⁴⁰ Subsequently, ‘in the event that the question should still arise whether an individual may rely on the directive as against a derogation laid down by national legislation,’ ⁴¹ in recalling the Becker judgement, the Court affirmed that the individuals may demand, vis-à-vis a State authority, the application of the principle of equal treatment between men and women as enshrined in Article 2, paragraph 1 of Directive 76/207, to the conditions of access to work and access to professional and further training, ‘in order to have a derogation from that principle under national legislation set aside’. ⁴² The effect, in the case, is direct because the application of EU law entails the disapplication of national law. Direct effect is objective in nature since the EU provision does not explicitly confer a right upon the individual and is ‘simply’ oppositive (rather than substitutive) since it is national law that finally applies, in conformity with EU law, although the latter is implemented at the level of a derogation enshrined therein.

The second ruling is Bernàldez, ⁴³ on Directive 72/166/EEC concerning the approximation of the laws of the Member States relating to insurance against civil liability in respect of the use of motor vehicles and enforcement of the obligation to insure against such liability. ⁴⁴ In the case at hand, the CJ urged the referring national judge to interpret the Directive as disapplying the Spanish provision whereby the insurer is not liable for damage to property and personal injuries caused to third parties, if the driver of the insured vehicle is intoxicated.

Two examples of the second strand of rulings are well-known cases. The first is CIA Security International, cited above, ⁴⁵ concerning Directive 83/189/EEC on the provision of information in the field of technical standards and regulations. In the ruling, the CJ observed that Articles 8 and 9 of the Directive – according to which Member States are required to notify to the Commission any draft of technical regulations falling under the sphere of application of the Directive – were sufficiently precise and unconditional to be invoked by the individuals before a national judge ‘which must decline to apply a national technical regulation which has not been notified in accordance with the directive’. ⁴⁶ As a matter of fact, the Directive contains provisions of a procedural character that can be relied on by individuals to obtain the freezing of domestic technical rules − that should have had been notified but were not − and do not directly confer any individual right because they ‘simply’ entail oppositive effects. In other words, there is no substitutive effect since EU law does not replace domestic legislation.

The second example, with regard to Article 9 of Directive 83/189/EEC, is the Unilever judgement, in which neither direct effect nor its conditions are mentioned. The Court’s assertion that Directive 83/189 ‘creates neither rights nor obligations for individuals’ ⁴⁷ has been interpreted in the sense of denying the direct effect of the provisions enshrined in the Directive. ⁴⁸ This denial holds true only if one considers direct effect as always linked to the creation of a right. However, this is an assumption, not an argument: assuming that direct effect always depends on the conferral of rights begs the question of whether there can be direct effect otherwise, rather than answer it. Instead, if one supports, as this essay does, the idea of a possible dissociation between direct effect and creation of a right, building upon the arguments developed above, the reasoning in the Unilever case − which is based in toto on that put forward in the CIA Security International ruling − cannot be of any surprise. The invocation of Directive 83/189/EEC by the Unilever SpA company and its subsequent enforcement by Italian courts ultimately triggered the disapplication of domestic technical regulations in relation to the labelling of foodstuffs, thereby causing a disadvantage for the Central Food company and, simultaneously, an advantage for Unilever. The effect entailed by the application of unconditional EU provisions is direct inasmuch as it imposed upon national authorities the duty to apply ‘old’ domestic technical standards rather than ‘new’ technical rules in force at the time in Italy – which had not been notified to the European Commission, as they should have been, by Italian authorities. It is direct, yet it neither generated substitutive effects nor conferred immediate individual rights. ⁴⁹

B. The decoupling between direct effect and conferral of rights with substitutive effects in the literature

Broad vs narrow definition of direct effect

Having discussed the case law, I am going to examine the literature, knowing that, particularly in view of the evolution of the case law, several authors have adjusted their approach over time: Timmermans ⁵⁰ and de Búrca, ⁵¹ for example, have abandoned the strict view of direct effect which inspired some of their writings, refraining, in more recent works, from any reference to the existence of a structural link between direct effect and attribution of rights.

At the roots of a broad understanding of direct effect lies the reconstruction carried out in the seventies by Waelbroeck, who identified two distinct categories of direct effect: ‘effet direct positif ou immédiateté’ and the ‘effet direct simple ou négatif’. ⁵² Moreover, still in the seventies, Bleckmann, even more boldly than Waelbroeck, anticipating what the CJEU would have argued in Becker and its other case law, speculated about the existence of a broader form of direct effect not necessarily linked to the conferral of an individual right. While underlining that the various formulas employed by the CJ revolved around ‘effets directs’ and ‘effets immédiats’, Bleckmann acknowledged that ‘la Cour de justice emploie le terme d’effet direct dans le sens plus large d’applicabilité par le juge’. In this vein, he pointed out that ‘la création de droits est un simple cas d’application de cette notion plus large’. ⁵³

Building upon Bleckmann’s studies and the German scholarship of early nineties (which was the first to grasp and systematically reflect upon an objective dimension in direct effect), ⁵⁴ Prechal, in her key 1995 monograph, reviewed and updated in 2005, observed that the definition of direct effect ‘in terms of the creation of individual or subjective rights as understood in national law will often be, if not impossible, then rather artificial and, moreover, unnecessary’. ⁵⁵ In 1997, Ruffert, in an article grounded in his 1996 monograph (in German), noticed that it could be possible to infer from the EU case law that the CJEU built the ‘objective direct effect as a new form of direct effect […] whereas the ordinary ‘subjective’ direct effect should remain conditional upon the existence of an individual right’. ⁵⁶ Edward reflected upon the existence of an objective direct effect insofar as he deemed it useful ‘to adopt the German distinction between objective and subjective direct effect’. ⁵⁷

A few years later, van Gerven, recalling the German literature, referred to the trend of objectification characterising the EU case law and observed that the provision of a directive could be invoked to seek the disapplication of national provisions which are incompatible with the said directive ‘in themselves, meaning, objectively without reference to any specific right granted to the claimant under the directive’. ⁵⁸ In similar terms, more recently, Bobek ⁵⁹ has explicitly argued in favour of ‘objective justiciability’.

The arguments against decoupling: objective-oppositive effect is not direct effect

At the heart of the literature theorising a narrow concept of direct effect − limited to its subjective-substitutive dimension −, ⁶⁰ we find the works by Winter, who, in the early seventies, affirmed that ‘the direct effects and the creation of individual rights […] seem to be inseparable and to stand to each other in a Yin-Yang relationship’. ⁶¹ Generally, in the literature, with the few exceptions mentioned above, this connection is taken for granted or strongly emphasised. ⁶² According to Dumon, direct effect can be summarised in the direct determination of subjective legal positions deriving from an EU norm. The author wrote that ‘sont des dispositions directement applicables celles qui, dans l’ordre interne, déterminent la position juridique des sujets de droit’. ⁶³ In the words of Lenaerts and Corthaut, direct effect must be considered as ‘the technique which allows individuals to enforce a subjective right’. ⁶⁴ Von Danwitz claimed that the ‘conséquence juridique réelle de l’effet direct’ is the creation ‘d’un droit dans la personne du citoyen de l’Union’ invoking EU law.

There are some recurring arguments in this literature against a broad definition of direct effect.

First, the scholars affirm that, when EU law confers a right to the individual which he/she may then exercise before the national authorities, the provision can be considered directly effective and, consequently, fill the gap left by the (inapplicable) national law. According to Wathelet, only the invocabilité de substitution would integrate a true hypothesis of direct effect since ‘c’est la règle communautaire et uniquement elle que le juge national est invité à appliquer pour garantir le droit conféré par l’ordre juridique communautaire au justiciable’. ⁶⁵

Second, this literature holds that, in the case of the oppositive effect, the EU provision would create ‘limited effects’ which are dissociated from the ‘claimable rights’ ⁶⁶ and, for this reason, it could not have true ‘ricadute soggettive’ ⁶⁷ (subjective repercussions) on the individual. In short, the EU provision would not be suitable ‘à completer directement le patrimoine juridique de particuliers de droits subjectifs et/ou d’obligations’. ⁶⁸

Third, few commentators believe that an EU provision generating oppositive effects could not be applied directly, in concrete terms, by the national authorities and, particularly, by the national courts. For this reason, the oppositive effect should fall within the concept of indirect effect. In this vein, the notion of justiciability would be similar to the one allowing the individual to seek compensation for damages or the interpretation of national law in conformity with EU law. ⁶⁹

Fourth, some authors note that the disapplication, as a result of oppositive direct effect, is limited to cases involving directives concerned with public law duties of a technical or procedural kind. ⁷⁰

The counterarguments in favour of decoupling: objective-oppositive effect is a form of direct effect

The above arguments are not conclusive. Indeed, there are four reasons why direct effect should be intended in a broad sense as comprising an objective-oppositive dimension and why this form of direct effect could not be rejected per se.

The first is that justiciability, in its oppositive facet, is not in itself uncommon. Indeed, it is not new or unusual in domestic legal systems. The so-called invocabilité d’exclusion is an essential element, for instance, in the French, Belgian and Dutch legal orders. ⁷¹

The second reason is that relevant legal effects for individuals can also be entailed by EU provisions that do not directly attribute rights to natural and legal persons. ⁷² Indeed, the legal subjective effects stemming from an EU provision are not only the result of a right created by EU law: a directly effective provision does not necessarily generate one type of effect. Two situations are worth mentioning: a) any right is directly and immediately attributed whereby the individual invokes an EU provision to ‘defend’ himself/herself from judicial proceedings that started against him/her at the national level; b) any right is directly and immediately attributed whereby the individual relies on EU law as a parameter of legality for challenging a State’s action in the context of administrative proceedings. As observed by Hohfeld, alongside the concept of ‘right in the proper sense’ there lie other concepts, such as ‘privilege’, ‘power’, and ‘immunity’, ⁷³ which are part of the broader notion of interest.

Prechal managed to frame the problem correctly. First, she observed that ‘direct effect is a broader concept, in the sense that a directly effective provision of Community law may be relied upon for several purposes’. Second, she noted that ‘to equate the concept of direct effect with the creation of rights does not do justice to the diversity of the effects which directly effective provisions may produce’. ⁷⁴ As observed by Bleckmann, ⁷⁵ the remedy of the legality review is a form of direct effect. Moreover, Isaac wrote that direct effect is ‘la capacité à être une source de la légalité en vigeur dans l’ordre juridique national’. ⁷⁶

Ruffert proves a point, too, when he remarks that ‘the direct effect of a directive provision is not dependent on the existence of an individual right that should be enshrined in the protective content or intent of the provision’. ⁷⁷ Bobek’s understanding of direct effect can be shared in so far as the ‘granting of individual rights’ is not seen as being ‘in itself’ ‘one of the general conditions for direct effect’. ⁷⁸ In the same vein, Schütze correctly affirmed that ‘direct effect does not depend on a European norm granting a subjective right, but to the contrary, the subjective right is a result of a directly effective norm’. ⁷⁹

As a matter of fact, the conferral of a right is a possible consequence caused by a directly effective EU provision, not its precondition. When the effect is objective-oppositive, the justiciability of the EU provision serves the purpose of protecting the interest of an individual that can take shape only through its application and, therefore, the neutralisation of the conflicting domestic provisions. Thus, there exists an interest, which is decoupled with the ‘attribution of a benefit to individuals’ ⁸⁰ and is incapable of representing a direct source for the creation of a right. Along these lines, the EU provision acts as a ‘standard for reviewing the legality of Member State action’. ⁸¹ What the individual invokes before national courts is the EU provision which, although not directly creating individual rights, imposes obligations towards the States. The EU provision is invoked as a means of defending the individual interest deriving from its fair application, even if not linked to the existence of a right recognised therein. A reliance on EU law, therefore, aims at enforcing the obligation(s) laid down in the EU norm. ⁸² This means that the conferral of a right is not an indispensable element per se: at the heart of a directly effective EU provision always lies an obligation and, very likely, but not necessarily, also a right. In this respect, Advocate General Tesauro, in the Brasserie case, noted that ‘the obligations of the Member States and Community institutions are directed above all, in the system which the Community system has sought and sets out to be, to the creation of rights of individuals.’ ⁸³ ‘Above all’, not always. Direct effect implies that the EU provision creates an obligation upon the State and/or an individual, and a corresponding advantage for a non-State party, which is not necessarily an individual right in the strict sense. As clarified by Bengoetxea, ‘the analysis of the legal relation involved in direct effect should always start with the obligation, ie, an obligation on someone, to do or secure something to somebody in some given circumstances and by some time’. ⁸⁴

The third reason is as follows: holding that the immediate application of EU law occurs only when the effect is substitutive because the EU provision would be fully self-sufficient only in this case, assumes, tautologically, that the oppositive effect is a minor manifestation of internal effect, which lacks one of direct effect’s constitutive elements, ie, the replacement of national law with EU law. This is incorrect for, through oppositive effect, domestic provisions can be applied and govern the case at hand autonomously, as has been shown with respect to, among others, the Bernàldez and Johnston cases.

Moreover, it shall be remembered that even a directly effective norm conferring a right upon the individual and apt to replace a domestic provision rarely fully takes the place of the national law that is at odds with the EU legal order. Said in other words, an EU provision is rarely perfectly autonomous. Additional clarifications, at the level of national law, are normally necessary. ⁸⁵ Furthermore, an EU law provision does not generally produce identical effects throughout the legal orders of the Member States. ⁸⁶ Hence, it is impossible to mark a stark boundary between subjective and objective direct effect.

The fourth reason challenges the argument according to which an oppositive effect, combined with the disapplication of national law, would come into play only regarding directives containing provisions of a technical nature. In fact, as clarified above in this section, from the Becker judgement onwards the CJEU did recognise the existence of oppositive effects also in relation to EU legal acts deprived of such technical character.

In short, considering the twofold form of direct effect, both objective and subjective, Prechal’s stance shall be embraced insofar as she defines direct effect as ‘the obligation of a court or another authority to apply the relevant provision of Community law, either as a norm which governs the case at hand or as a standard for legal review’. ⁸⁷

The common element between substitutive-subjective direct effect and objective-oppositive direct effect is the occurrence of the exclusion, ie, the disapplication of national law. Moreover, when engaging with a directly effective EU provision, national authorities, being bound by the obligation of disapplication, always settle the case by virtue of a different provision than the one that would have applied if, ab astracto, EU law had not been applicable: in the case of subjective-substitutive direct effect, the EU provision instead of the national one; in the case of objective-oppositive direct effect, a national norm different from the one originally envisaged to govern that specific case. ⁸⁸ The fundamental divergence between substitutive-subjective direct effect and objective-oppositive direct effect, instead, is that while the exclusion is always present within subjective direct effect together with a substitutive effect because the EU provision directly creates a legal position governing the case at stake in place of the internal law provision, within objective direct effect the effect is ‘only’ oppositive since the dispute is governed by the internal law which outlives (and was not affected by) the disapplication.

4. Beyond the objective-oppositive/subjective-substitutive dichotomy: at the roots of disapplication

A. Direct effect and primacy (not primacy alone) as a trigger of disapplication

In addition to the points raised in section 3 about the possible decoupling between direct effect and the creation of rights, among the arguments against and those in favour of objective-oppositive direct effect lie others concerning the relationship between direct effect, primacy and disapplication. The reflection upon a narrow or broad conception of direct effect, therefore, is an occasion to venture, in this section, into the role and scope of the remedy of disapplication, in its interplay with the principles of direct effect and primacy.

Several Advocates General ⁸⁹ and authors ⁹⁰ have argued that disapplication is not always the product of both primacy and direct effect. They claim that in the CJEU case law on objective-oppositive effect only the principle of primacy would apply and cause the disapplication of domestic provisions. ⁹¹ The oppositive effect would simply act as a ‘corollary’ of primacy. ⁹² As noted by Advocate General Léger in the Linster case, where one of the parties did not aim to obtain recognition of an individual right, the matter of direct effect ‘tends to be eclipsed by that of primacy’ because the problem of integration of the EU provision in the national legal order would remain ‘a confrontation between the different rules’ while ignoring the rule regarding the application of the said norms to the subjects of rights. ⁹³ The result is a real disconnection between direct effect and primacy as far as disapplication is concerned. This dissociation becomes possible as direct effect is deemed to be a sufficient condition ‘pour entraîner le double effet d’exclusion et de substitution, mais que rien ne permet d’affirmer qu’il s’agit d’une condition nécessaire’. ⁹⁴ Moreover, it was affirmed that ‘pour écarter une norme nationale contraire, point n’est besoin, pour le juge national, de lui reconnaitre un effet direct’. ⁹⁵

In this regard, one could argue that only an outdated reading of the Simmenthal ⁹⁶ judgement may lead to denying the disapplication of domestic law in situations that are not ‘covered’ by EU provisions endowed with direct effect. In the same vein, it may also be observed that the recognition of direct effect as a precondition for disapplication could dangerously restrict the effet utile of EU law, with the additional risk of improperly expanding the scope of the remedies of consistent interpretation and State liability, as well as the constitutional review mechanism for States, such as Italy, that foresee it.

Furthermore, several scholars ⁹⁷ and Advocates General ⁹⁸ have argued that tracing back the oppositive effect to direct effect would call into question the axiom of EU law recognised from the Faccini Dori ruling ⁹⁹ onwards, according to which directives, being addressed at States, could not be intended as prescribing obligations towards individuals and, consequently, could not be invoked against them by other private parties. In other words, part of the literature highlights that tracing back the objective-oppositive effect, and disapplication, to primacy alone, rather than to direct effect and primacy, would solve the problems arising from a number of cases, including the above-mentioned CIA Security International and Unilever judgements. Indeed, in this case law the disapplication entailed by EU directives affects, de facto and in incidental/indirect terms, the private individuals who had benefited from the Member State’s ‘guilty’ omission. The lack of directives’ horizontal direct effect would then be infringed because of oppositive direct effect. On the contrary, scholars have affirmed that the exclusion of oppositive effect from direct effect would make such infringement a priori impossible. According to Advocate General Bot in his Opinion in the Kücükdeveci case, ‘disconnecting the horizontal direct effect of directives from the right to plead them to exclude contrary national law’, would constitute a ‘palliative’ for the lack of direct effect, whereby the application of an EU directive would cause adverse effects for the individuals in the horizontal relationships. ¹⁰⁰

The reasoning above raises criticism for the following reasons.

First of all, a comprehensive concept of direct effect is preferable to a strict one when it comes to the matter of directives’ internal effect in horizontal relationships, because the effect, in such situations, is not properly horizontal. In fact, at the core of the individual’s action stands the violation of an obligation addressed to the State, rather than simply the behaviour of a private party. ¹⁰¹ And, in any case, as pointed out by the majority of scholars engaging with the unjustified constraints and incongruencies implied in the prohibition of directives’ horizontal direct effect, ¹⁰² the hope is that, in the future, the CJEU will change its attitude and recognise the right, for individuals, to enforce provisions contained in a Directive also in horizontal relationships. ¹⁰³

Most importantly, the primacy principle would be the only trigger of disapplication if one were to embrace a strict interpretation of direct effect. ¹⁰⁴ Such stance, however, as explained in section 3, shall be rejected.

From the standpoint of the CJEU case law, historically, the case law has been ambiguous because the EU courts have not extensively dealt with the nexus between direct effect, primacy and disapplication altogether. Notwithstanding such uncertainties, it seems to me that three landmark rulings show that direct effect is always the condicio sine qua non for triggering the obligation to disapply national laws: Simmenthal, Popławski II and NE v Bezirkshauptmannschaft Hartberg-Fürstenfeld II. ¹⁰⁵

In Simmenthal it was, first, affirmed that ‘the relationship between provisions of the Treaty and directly applicable measures of the institutions on the one hand and the national law of the Member States on the other is such that those provisions and measures […] by their entry into force render automatically inapplicable any conflicting provision of current national law’. Secondly, it was observed that ‘every national court must, in a case within its jurisdiction, apply Community law in its entirety and protect rights which the latter confers on individuals and must accordingly set aside any provision of national law which the latter confers on individuals’. ¹⁰⁶ Notwithstanding the confusion that may arise due to the reference to the concept of direct applicability, rather than direct effect, and the lack of an explicit link between direct effect and the remedy of disapplication, it can be easily drawn from the judgement that disapplication, while being a manifestation of primacy, is triggered by the application of EU directly applicable/effective provisions. On the other hand, it shall be stressed that direct effect is not a precondition, and primacy applies alone only in respect to the effect of preclusion/pre-emption, ie, whereby EU provisions ‘preclude the valid adoption of new national legislative measures to the extent to which they would be incompatible with Community provisions’. ¹⁰⁷

Recently, the qualification of direct effect as a trigger for imposing disapplication has been challenged, for the first and only time in explicit terms, by the CJ in the Link Logistic ruling, ¹⁰⁸ which followed, in this respect, the opinion by AG Bobek. ¹⁰⁹ More specifically, the EU judges, while excluding direct effect in respect to the principle of proportionality of penalties in Article 9a of Directive 1999/62, ¹¹⁰ affirmed that national authorities shall interpret national law in conformity with EU law. In this vein, the CJ maintained that, ‘if such an interpretation is not possible’, domestic authorities shall ‘disapply any national provision in so far as its application would, in the circumstances of the case, lead to a result contrary to EU law’. ¹¹¹ Such statement is stunning and unprecedented since it stands in contrast with the Simmenthal ruling. Moreover, the reference to the case law cited by the CJ in Link Logistic in support of its vision of direct effect is misleading. For instance, the Pöpperl judgement that the EU judges mentioned twice in their ruling, ¹¹² in which an obligation was imposed upon domestic courts to disapply national law if consistent interpretation is not feasible, does not concern at all, unlike the Link Logistic ruling, EU non-directly effective provisions.

A few weeks later, a similar position was adopted by AG Sánchez-Bordona in his opinion rendered in the Popławski II judgement. ¹¹³ However, this stance has been bluntly rejected by the CJ in Popławski II. As clarified by AG Bobek in the NE v Bezirkshauptmannschaft Hartberg-Fürstenfeld II case, ¹¹⁴ the Popławski II ruling has ‘authoritatively settled the issue of whether the setting aside of national law may be only the consequence of primacy, or of primacy of EU law in conjunction with the direct effect of the EU law provision that is to be applied’. Indeed, with the Popławski II judgement a firm and definitive point has been made regarding the interplay among direct effect, primacy, and disapplication. The CJ boldly rejected the argument according to which primacy could and should autonomously (without direct effect) require national authorities to disapply. A further consequence is that the EU judges, albeit implicitly, cut to the chase with regard to the characterisation of the objective-oppositive effect: it is direct effect, and there is no doubt about it, since it triggers the disapplication of conflicting domestic law. ¹¹⁵ In particular, it was stated that the principle of primacy cannot ‘have the effect of undermining the essential distinction between provisions of EU law which have direct effect and those which do not and, consequently, of creating a single set of rules for the application of all the provisions of EU law by the national courts’. ¹¹⁶ Moreover, the CJ affirmed that ‘a provision of EU law which does not have direct effect may not be relied on, as such, in a dispute coming under EU law in order to disapply a provision of national law that conflicts with it’. ¹¹⁷ Finally, the EU judges observed that ‘a national court is not required, solely on the basis of EU law, to disapply a provision of its national law which is contrary to a provision of EU law if the latter provision does not have direct effect’. ¹¹⁸ Consequently, the conclusion stemming from the Popławski II ruling can be summarised as follows: just as the Simmenthal judgement shall be read in the sense of imposing upon national authorities an obligation of disapplication when the EU provision is directly effective, in the same way it shall be read in the sense of not requiring the disapplication of a national norm when an EU provision deprived of direct effect is at stake. A domestic judge, therefore, may decide to disapply even when direct effect does not arise, but in this case the concerned national legal order should make sure that the legal positions of those subjects affected by disapplication are sufficiently safeguarded, although EU law, lacking direct effect, cannot govern the case. The recent judgment delivered by the CJ in the Thelen Technopark case, which has been curiously overestimated particularly by one author in its scope, ¹¹⁹ is a confirmation of this reasoning. By the judgement the CJ did not impose disapplication on the basis of an EU provision lacking direct effect; what it did was simply clarifying that national authorities can (not must) ‘disapply, on the basis of domestic law, any provision of national law which is contrary to a provision of EU law that does not have such effect’. ¹²⁰

The NE v Bezirkshauptmannschaft Hartberg-Fürstenfeld II judgement shall be considered too. Indeed, in this decision, the CJ, by explicitly overruling its previous Link Logistik judgement, affirmed that, contrary to what was held in paragraph 56 of such judgement, the requirement of proportionality of penalties laid down in Article 20 of Directive 2014/67 ¹²¹ is unconditional and sufficiently precise ‘to be capable of being invoked by an individual and applied by the national administrative authorities and courts’ in order to challenge domestic legislation providing for disproportionate penalties. ¹²² This is the true and only reason why disapplication could come into play, differently from what was stated in Link Logistik, whereby, as clarified above, EU judges required national authorities to set aside domestic legislation, albeit being the principle of proportionality of penalties considered deprived of direct effect.

In this connection, we shall conclude that if the principle of direct effect presupposes, to be effective, the primacy of EU law, the reverse is not true. ¹²³ The argument according to which the obligation to disapply could be a result of primacy alone forgets that the primacy principle has a larger sphere, when compared to direct effect, insofar as also non-directly effective provisions, as known, impose obligations towards national authorities. As a matter of fact, EU provisions are not per se directly effective: while they all prevail over the internal legislation of the Member States, only some of them are directly effective. This is the summa divisio that was conceptualised in the literature decades ago and still stands today. ¹²⁴

Against this background, why shall direct effect be considered the precondition for disapplication, as the result of the application of EU law?

Direct effect ensures that the UE principle of conferral, and the related principle of subsidiarity, are not called into question by a generalised, unrestrained use of the principle of primacy. Justifying disapplication on the basis of primacy alone would be dangerous and potentially fatal for the whole dialogue between EU institutions and Member States. It would entail a blank proxy capable of undermining the fertile relationship between EU law and domestic legal systems for good, as well as the mutual cooperation between EU institutions and Member States’ authorities. ¹²⁵

Arguing that the obligation to disapply cannot arise outside the realm of direct effect, whether it be substitutive or oppositive, does not mean, of course, that consequences stemming from the application of the primacy principle shall not arise. On the contrary, besides the avenue provided by Article 267 TFEU, individuals and national authorities − including judges − have at their disposal the remedies offered by the combination of the EU and national legal orders: consistent interpretation, if the interpretative tool can be effectively pursued; State liability; and the referral to domestic constitutional/supreme courts for an erga omnes decision on the incompatibility of national law with the EU legal system.

B. When can, or must, disapplication be precluded, suspended or postponed, notwithstanding the occurrence of a conflict between national law and an EU directly effective provision?

While disapplication, as an obligation stemming from EU law, should always take place in the context of directly effective EU provisions as a result of both direct effect and primacy, it is not always true that direct effect of an EU provision implies necessarily, automatically and immediately the disapplication of the conflicting domestic norm.

In this respect, my understanding of the EU case law is that four exceptions arise. These are legitimate, physiological derogations from the principle of primacy since the CJEU itself finds them, in principle, compatible with the EU legal system. Thus, they fall outside both the well-known counterlimits doctrine and the Czech, ¹²⁶ Danish, ¹²⁷ German ¹²⁸ judicial rebellions occurred in the recent years with regard to, respectively, the role of the European Central Bank, the horizontality of general principles of EU law and the scope of EU primary and secondary law in the field of social security. Moreover, they fall outside the recent confrontational dialogue regarding the rule of law, as it is safeguarded in Articles 2 and 19 TEU, undertaken by the CJEU with Hungarian, Polish and Romanian Governments and constitutional/supreme courts. ¹²⁹ The latter are different, pathological situations since the arguments invoked by Member States indicate the existence of a conflict between EU law and national legal orders which is impossible to recompose.

A first exception arises whereby the CJEU, despite the directly effective nature of one or more EU provisions, chooses to refer the matter to the national judge to achieve a fair balance of interests implied in the enforcement of EU law. This means that the bond between direct effect and immediate disapplication breaks when EU judges, even though confronted with EU directly effective provisions, refrain from directly conducting a balancing of interests that they deem falling instead within the competence of the national court submitting the reference for a preliminary ruling. An example is the Österreichischer Rundfunk case, whereby the CJ, while stating that the articles of Directive n. 95/46/EC at stake were directly effective and thus ‘may be relied on by an individual before the national courts’, concluded that it was ‘up to the national judge’ to verify, pursuant to the principle of proportionality, whether it was necessary to immediately oust the application of domestic law. ¹³⁰ In this respect, when the matter is the protection of fundamental rights, should the CJEU refuse to issue an erga omnes interpretation of EU law that would lead to the disapplication of national law, the national judge could, or even should, seek an erga omnes ruling by referring the matter to the national Constitutional Court in systems that foresee such mechanism, as was highlighted in Italy by the Constitutional Court in its ruling n. 20/2019. ¹³¹ In these types of proceedings, the decoupling of direct effect and disapplication is viable due to a previous non liquet coming from the CJEU.

A second exception arises whereby the CJEU recognises that ordinary judges can exceptionally and temporarily suspend the ousting effect which an EU directly effective provision has on national law in so far as a number of conditions are fulfilled. First, the national measure whose disapplication is excluded is a measure that correctly transposes a directive. Second, the adoption and entry into force of the new national measure would not enable the adverse effects resulting from the disapplication of the contested measure to be avoided. Third, the legal vacuum as result of disapplication would result in the attainment of objectives which would run specifically counter to the fundamental objective of the directive. Fourth, the effects produced by the contested measure are exceptionally maintained only for the period of time which is strictly necessary to adopt the measures apt to undo the irregularity that led to the emergence of a contrast with the EU legal order.

Now, this derogatory rule of reason has been identified, framed and admitted, to date, only in the field of environment and energy supply, whereby the CJ relied on ‘imperative reasons of overriding public interest’ as the only trigger for deeming legitimate the suspension of the remedy of disapplication. ¹³² On the other hand, it can be inferred from the whole reasoning developed by the CJ in the Winner Wetten GmbH case ¹³³ that the provisional suspension of disapplication is not limited to only clear-cut and limited circumstances, provided that ‘imperative reasons of overriding public interest’ occur. ¹³⁴ This understanding is confirmed in the Inter-Environnement Wallonie and Bond Beter Leefmilieu Vlaanderen ruling, in so far as the CJ explicitly observed that the derogation from disapplication comes into play ‘on a case-by-case basis […] with due regard to the conditions laid down by the Court’s case-law’. ¹³⁵ In any event, the provisional suspension of disapplication should be confined to situations of an exceptional nature, considering the repercussions that it might have on the scope of the primacy principle.

A third exception arises whereby the CJEU, on one hand, does not find room for disapplication, albeit being the EU law at stake directly effective, on the other hand, imposes an obligation upon ordinary judges to extend temporarily to individuals − pending an ad hoc domestic provision enabling to remove the incompatibility of national law with EU law − rights and advantages that national law already grants to specific groups of subjects. This exception occurs specifically in the field of equality and non-discrimination, as shown by the Cresco Investigation case, ¹³⁶ in which the disapplication of a national legislation granting certain employees (with a particular religious affiliation) a supplementary day’s holiday − that the Court found discriminatory pursuant to Article 21(1) of the Charter − would have entailed a generalised reduction of guarantees for all individuals, while not attributing any benefit to the disadvantanged group of employees who belonged to other churches and were not included in the scope ratione personae of the domestic provision whose disapplication had been suspended by the CJ. ¹³⁷

A fourth exception arises when the CJEU approves the invocability of the national identity clause foreseen in Article 4(2) TEU ¹³⁸ by the State involved in the proceeding before the EU judges, which normally takes place in the context of Article 267 TFEU and is often grounded in the position taken by the Constitutional/Supreme Court in its jurisprudence. If this is the case, the principle of primacy is derogated and disapplication cannot be performed, regardless of the directly effective character of EU law. Of course, as affirmed mildly in Torresi ¹³⁹ and Coman ¹⁴⁰ and bluntly in the RS ¹⁴¹ ruling, such derogation can be admitted only by the CJEU since it is the latter the ultimate interpreter of a Treaty provision such as Article 4(2) TEU, in spite of the renvoi provided therein to national identities, ie, first and foremost to constitutional legal orders of Member States. This reconstruction of the national identity clause shows to what extent the process of European integration developed in quasi federal terms, with two major implications. First: the CJEU, in attracting in its exclusive competence Article 4(2) TEU, as parameter of legality of EU law, through Article 267 TFEU and the preliminary reference proceeding on validity, inevitably puts in the center of its analysis the domestic legal order and its validity, thus expanding the scope of the preliminary ruling. A ruling which, as known, being erga omnes, should in principle concern the interpretation and validity of EU law, not of national law. Second: the CJEU, by virtue of a provision, such as Article 4(2) TEU, that refers to (all) the domestic legal orders of Member States, gets to be the only interpreter, not only of EU law, but also of the 27 national legal systems, insofar as their national/constitutional identities come into conflict with the principle of primacy. That way, the CJEU takes on new responsibilities even more than in the past, being the concept of national identity naturally evolutive and changeable.

5. Clarity, precision, and unconditionality: the test of direct effect revisited

A. The test’s premises

As is well-known and has been mentioned in section 2, the CJEU, in its case law, ¹⁴² generally identifies the following conditions for direct effect: clarity, precision, and unconditionality. The obligation enshrined in the EU provision must be sufficiently determined and complete, ie, ‘legally perfect’, ¹⁴³ so that the individuals may piece together its content and invoke the EU norm before national authorities. Should the norm be directly effective, such authorities (public administration and local jurisdictions), in turn, shall be obliged to apply EU law and, where necessary, disapply the conflicting domestic law provisions.

Regarding clarity and precision, the CJ believed them to be integrated once the provision ‘sets out an obligation in unequivocal terms’. ¹⁴⁴ To date, what was noted in the Francovich judgement still holds true. In Francovich it was stated that an EU provision is clear and precise if there exist, cumulatively, three elements (alternatively, they may be derived as the result of interpretation): the determination of the person(s) entitled to benefit from the provision, the determination of the person(s) obliged to grant the right/advantage implied therein, and the content of such right/advantage. ¹⁴⁵ Individuals must be able to ascertain what their rights and/or obligations are, ¹⁴⁶ and national authorities must be able to apply the provision(s) to the case at hand.

With respect to the prerequisite of unconditionality, according to the CJEU, whether the application of the EU provision is subject ‘in its implementation or effects’ ¹⁴⁷ to the issuing of additional integrative and/or implementing measures by the national authorities or the EU institutions must be assessed. ¹⁴⁸ As noted by the EU judges, the existence of a margin of discretion does not necessarily imply that the provision is per se deprived of direct effect. Indeed, it is true that unconditionality is not an absolute concept and that an EU provision is directly effective also when Member States enjoy a certain degree of latitude in enforcing EU rules. ¹⁴⁹

B. The test’s outdatedness

I believe that the questions one should ask when confronting the scope and extent of the test on direct effect, from the standpoint of national courts charged with the application of EU law, ¹⁵⁰ are as follows: are precision, clarity, and unconditionality genuinely used by the CJEU to verify the direct effect of EU law? Do they have the same relevance in the case law? Does their assessment vary due to the characteristics of the EU provision(s) at stake? Are they still, in practice, useful for an appropriate understanding of direct effect? In the next sections I will answer those questions and explain why I consider the test ultimately outdated and ineffective.

The test’s variables: precision and clarity

The test, in itself, is outdated for five main reasons. The first three will be discussed in this section, while the last two in the next one.

Firstly, nowadays, in many rulings, the concept of clarity, which was normally referred to in past case law, ¹⁵¹ is not even recalled. Indeed, clarity and precision have the same meaning: using both terms, as if they were two distinct premises, is redundant and misleading. ¹⁵²

Secondly, the interpretation of the conditions of direct effect made by the CJEU is usually extensive. ¹⁵³ Provisions which, upon an initial reading, are not clear, precise, or unconditional, have been considered as such by the EU judges through a flexible and teleological interpretation. Such an approach is evident in the Van Gend & Loos judgement, and is even more so in the subsequent case law, starting with the Reyners ¹⁵⁴ and Defrenne ¹⁵⁵ cases, where the test was further ‘relaxed’, ¹⁵⁶ ‘loosened’ ¹⁵⁷ and ‘broadened’ ¹⁵⁸ by the CJEU so as to avoid that even less accomplished provisions than Article 12 TEC could not satisfy it. Furthermore, not only is the interpretation of the three criteria for direct effect generally very broad, but even the lines between them became very thin. As Advocate General van Gerven observed in his opinion in the Banks case concerning the criteria for direct effect, ‘on closer scrutiny, the case law of the Court exhibits several minor differences as regards the wording of those conditions’. ¹⁵⁹

Thirdly, the fact that the CJEU, in its case law, often verifies the presence of precision, clarity, and unconditionality does not mean that it always does so and with reference to all sources of EU law. ¹⁶⁰

With regard to negative obligations, at times the test has been blandly applied or not even mentioned. The EU Court’s simplistic affirmation in the Molkerei-Zentrale Westfalen/Lippe GmbH case that EU provisions which contain a prohibition are ‘by their nature’ ¹⁶¹ sufficiently precise, unconditional, and, therefore, directly effective, still stands today.

In respect of positive obligations, the assessment on the conditions of direct effect generally tends to play a more decisive role since these are provisions which confer a subjective legal position enabling the individual to obtain a benefit vis-à-vis the State (or private subjects).

In this framework, it is worth mentioning the different attitude shown by the EU judges vis-à-vis the assessment on direct effect’s conditions, depending on whether it is of the substitutive or oppositive type. Indeed, it seems to me that, where the EU norm penetrates the national legal order, substituting itself for the pre-existing domestic provision, the test is generally carried out, while the same cannot be said in relation to those EU provisions which do not directly confer subjective legal positions and that aim at avoiding, in oppositive terms, the application of conflicting internal norms. ¹⁶² By way of example, in the Verbond, ¹⁶³ ENKA, ¹⁶⁴ Delkvist, ¹⁶⁵ Cowan, ¹⁶⁶ Kraaiejveld, ¹⁶⁷ WWF, ¹⁶⁸ Linster, ¹⁶⁹ Unilever, ¹⁷⁰ and Landelijke ¹⁷¹ cases, the test is not even mentioned. The reason seems to be the necessity, for the CJEU, the legal interpreter, and the national operators, to have better ‘guidance’ in the cases in which the external (EU) provision takes the place of the internal one, as happens with substitutive direct effect, rather than what generally happens where said provision is invoked by the individual ‘only’ as a parameter of legality of national law (oppositive direct effect). ¹⁷²

As observed by Advocate General Léger in the Linster case, the limits regarding the role of the legislator when the effect is oppositive ‘are not the same when it is called upon to assess the validity of a rule of domestic law in the light of the relevant Community rule’. ¹⁷³ This does not always occur. In the CIA Security International case, with relation to the oppositive effect, par excellence, of an EU provision, ¹⁷⁴ for example, it is explicitly observed that the provisions of Directive 83/189/EEC are unconditional and sufficiently precise. ¹⁷⁵

In general, it is my impression that the Court has applied the three criteria vaguely and in a shifting, evanescent manner, as if following a variable geometric approach, without any clear guiding principles. At times, it has applied these criteria, but without making any mention of direct effect (eg, WWF, El Dridi). At others, it has discussed direct effect but without making reference to the criteria (eg, Verbond). In a large number of cases, the Court has made no mention of either direct effect or the criteria (eg, Melki, Åkerberg Fransson, Bernáldez, Unilever). This absence leads to a reassessment/relativisation of the three premises. The foregoing applies to the EU legal order as a whole, as is clearly confirmed by the case law on the general principles of EU law and the Charter of Fundamental Rights. In fact, in most of the CJEU rulings on the Charter ¹⁷⁶ a reference to the conditions of direct effect and/or the test itself is lacking.

The test’s only and constant condition: unconditionality/direct applicability

Fourthly, the criterion of unconditionality, which at its core is an assessment of the margin of discretion left to national authorities when applying the EU provision, tends to absorb the condition of precision. In other words, if a norm is clear/precise, it is not necessarily unconditional, while if a norm is unconditional, it is inevitably clear/precise. As a matter of fact, if there is no need for the adoption of a further internal legal act to execute the EU provision, then the obligation contained therein, due to its unconditional character, is also in re ipsa clear/precise.

Fifthly, just as direct effect/applicability of regulations does not a priori exclude that provisions contained therein should be implemented, through further acts, at the domestic level, ¹⁷⁷ other EU legal sources, both primary and secondary, including directives, although not defined as directly applicable by the Treaties or the CJEU, may contain obligations which do not require the adoption of further measures. ¹⁷⁸ Aside from issues of terminology, direct applicability is a feature of regulations, but it can also be part of other EU legal sources, under the guise of their unconditional character, including directives. There are no significant differences between the effects of regulations and unimplemented directives when further intermediate national acts are not needed. ¹⁷⁹ In such situations direct effect arises and, of consequence, does direct applicability.

More in particular, as to the interplay between direct effect and direct applicability, scholars tend to generally exclude the possibility that these two concepts may coincide. The main idea is that, while direct applicability refers to an objective quality of an EU provision not requiring a further implementing act, direct effect, instead, is about the impact of EU law on the individual through the conferral of a right. As noted in the literature, there is a need to distinguish between direct applicability and direct effect because the former indicates the way EU law is implemented in the legal order of the Member States as the ‘law of the land’, while the latter is concerned with how said ‘law of the land’ can be invoked by private subjects before national courts. ¹⁸⁰

Such reconstruction is theoretically correct. However, the problem regarding the relationship between direct effect and direct applicability must also be considered from a more practical point of view. The evidence demonstrates that direct effect and direct applicability cannot be deemed as distinct legal categories for two reasons. The first is that, from an analysis of EU and national case law, it appears that direct effect and direct applicability often merge: what really matters, in practice, ¹⁸¹ is whether an EU provision is justiciable, ie, applicable to the dispute at issue. For instance, in the Reyners ruling, regarding the (then) Article 52 TEEC, now Article 49 TFEU, concerning the freedom of establishment, the norm was defined as ‘directly applicable’/‘directement applicable’, ¹⁸² even though the invocation of a provision by the individual was at stake rather than the lack of need for an act of implementation.

The second, crucial reason is as follows: if unconditionality means direct applicability and if direct applicability/unconditionality, by ‘absorbing’ the conditions of clarity and precision, is the only core element of (and precondition for) direct effect, direct effect and direct applicability end up meaning the same thing rather than being ‘simply’ interconnected. ¹⁸³

6. The true conditions for direct effect: unconditionality and the creation of an advantage for the individual

In light of the above considerations, when can an EU provision actually be considered directly effective? The presence of two cumulative requirements must be verified by the national judge. The first is that the EU norm is unconditional and thus immediately/directly applicable. In line with Pescatore, direct effect shall be identified with the concept of justiciability of a norm ¹⁸⁴ and with its ability to be directly applied. In the words of Edward, the doctrine of direct effect, ‘using that expression in a broad way’, provides the criteria ‘for selecting or rejecting the norms to be applied and for clarifying the scope of judicial competence’. ¹⁸⁵ Hence, at the heart of direct effect lies the concept of immédiateté, ¹⁸⁶ that is the capacity of an EU provision to govern the case at hand, with subjective-substitutive or objective-oppositive effects. As proof of the practical nature of direct effect, ¹⁸⁷ Advocate General Van Gerven correctly affirmed that, ‘provided and in so far as a provision of [Union] law is sufficiently operational in itself to be applied by a court, it has direct effect’, ¹⁸⁸ and, in the same vein, more recently, Advocate General Bobek observed that ‘direct effect is the ability of an EU law rule to be justiciable at national level, directly before the national judge, without the need for any further “intermediary” of national law’. ¹⁸⁹

There exists a second condition which is generally neglected in the literature: the application of the EU provision in lieu of the national norm (subjective-substitutive direct effect), or as a parameter of legality of the national norm (objective-oppositive direct effect), must create an advantage for the individual. This means that the doctrine of direct effect, which is conceived as a means at the disposal of natural and legal persons to actively safeguard their own prerogatives and ultimately enforce EU law, cannot entail only a disadvantage for the individual. The minimum requirement for the emergence of direct effect is the existence of an advantage/interest, ¹⁹⁰ which occurs as a result of the application of EU law and of the subsequent disapplication of domestic law, regardless of whether said position/interest is (subjective-substitutive direct effect) or is not (objective-oppositive direct effect) conferred directly by the same EU norm and regardless of whether − in addition to the benefit for the individual − there simultaneously lies a constraint of personal prerogatives for someone else, as happens in the context of horizontal direct effect. In short, if the enforcement of an EU provision is entirely in malam partem, lacking the creation of an advantage for anybody, direct effect does not arise and, therefore, disapplication cannot operate. ¹⁹¹ This shall apply not only in the context of criminal law, as stated by the CJ in a number of cases, ¹⁹² and not only with regard to directives, which cannot, as known, entail direct effects in horizontal relationships. ¹⁹³ Indeed, the prohibition of in malam partem application of EU law − and of the related disapplication of national law − shall stand in all areas of law and with regard to the EU legal system as a whole, including primary law, the Charter and secondary law beyond directives. This means that direct effect is not only a quality of the provision; in fact, it should be assessed in light of the consequences for the individuals stemming from the application of EU law and the disapplication of national law. Such attention to the outcome of the application of the EU rule in each individual case has been recently acknowledged by Advocate General Bobek in NE in respect to the principle of proportionality of penalties, yet it can be transferred into other realms of the EU legal system. ¹⁹⁴ Contextualising direct effect means taking into consideration the implications arising from the enforcement of EU law in the national legal system, with respect to the case at hand, from the standpoint of the effects resulting from the disapplication of the conflicting domestic norm.

Therefore, the EU judges make a mistake when they impose the disapplication of a national provision upon domestic judges on the basis of an EU provision that is not unconditional or, albeit unconditional, results only in harming the interests of the individual. Of course, it might be claimed that, while the disapplication of national law would entail in malam partem effects in the dispute examined by the domestic judge for both the applicant and the defendant, it would also generate advantages for the State and for the community at large, being aimed at pursuing general interest objectives. This is true. However, whereby the enforcement of EU law affects the rights of all parties to the dispute, without creating any advantage for the latter, the principle of legitimate expectation and legal certainty turn out to be necessarily infringed. In other words, the occurrence of an advantage for the State or for natural and/or legal persons different from the parties to the dispute is not relevant for the purposes of affirming that, in such situations, direct effect shall arise. Advocating that direct effect could come into play regardless of the creation of an advantage for the individual for such doctrine was born also for ensuring the private enforcement of EU law, in addition to the private enforcement by the European Commission pursuant to Article 267 TFEU, ¹⁹⁵ is misleading. Indeed, one thing is to acknowledge that direct effect is a tool used by the individual to safeguard his/her rights (and seek redress) and thus enforce EU law, another is to claim that direct effect may serve, in practice, only as a means to apply EU law, while diminishing rights for all parties to the dispute. As a matter of fact, should there be a clear conflict between EU and national law, unsolvable by virtue of (consistent) interpretation, the State has other means (than disapplication), provided for at the national level, for complying with EU law and enforcing the principle of primacy, while respecting the principles of legality and legitimate expectations. Most notably, in those countries equipped with a centralised system of constitutional review, the remedy is represented by a referral to the Constitutional Court, which would then render an erga omnes judgement striking down the conflicting domestic provision(s). Hence, a mechanism similar to the one envisaged for the enforcement of the European Convention of Human Rights’ provisions, as well as of the European Court of Human Rights’ jurisprudence, exists. ¹⁹⁶ Conversely, in those countries lacking a centralised system of constitutional review, the antinomy would be resolved through the involvement of their respective Supreme Courts. ¹⁹⁷

Two well-known examples can be drawn from the Italian experience. Notably, the chaos that was generated amongst judges, practitioners, and scholars in Italy regarding the Lucchini ¹⁹⁸ and Taricco sagas ¹⁹⁹ − and the consequences due to the infringement of constitutional principles resulting from the CJEU’s reasoning − could have been avoided if the CJ had, in both cases, denied the directly effective character of the EU provisions at stake, instead of (more or less explicitly) arguing otherwise.

With respect to Lucchini, at issue was the understanding of whether EU law precluded the application of Article 2909 of the Italian Civil Code because such rule prevents the recovery of a State aid that was granted in contrast with Article 107 TFEU and which incompatibility with the common market was declared in a Decision of the Commission. The CJ, in its ruling, overturned the principle of intangibility implied in the res judicata, observing that Article 2909 hinders the enforcement of EU law ‘in so far as it would make it impossible to recover State aid that was granted in breach of Community law’. ²⁰⁰

With regard to Taricco, at stake was the possible disapplication of Article 160, last paragraph, and Article 161, second paragraph, of the Italian Criminal Code, in so far as such provisions foresaw narrow limitation periods for the prosecution of serious VAT frauds and, for this reason, were liable to have an adverse effect on the fulfilment of Member States’ obligations under Article 325 TFEU. While in Taricco I the CJ observed that the domestic provisions at stake were suitable to affect the obligations imposed to the Member States by Article 325, paragraphs 1 and 2, TFEU, ²⁰¹ in Taricco II − in the context of a preliminary reference by the Italian Constitutional Court ²⁰² − it specified that national jurisdictions had to verify whether the judiciary-based ‘Taricco rule’ elaborated by the CJ could violate the principle of legality of criminal offences and penalties ‘in its preconditions of foreseeability, specificity and non-retroactivity of the applicable criminal law’. ²⁰³ The Italian Constitutional Court, in response to the CJ’s Taricco II judgement, by de facto invoking the counterlimits doctrine, concluded that Italian judges were not held to apply the Taricco rule, and, therefore, ‘Article 160 last paragraph, and Article 161 of the criminal code remained applicable’. ²⁰⁴

If direct effect were not at the origin of the stance taken by the CJ, all problems relating to the impact on national constitutional principles as a result of disapplication could have been circumvented. Indeed, direct effect should not have had a role in either Lucchini nor in Taricco since the application of EU law in both cases harms the individual’s interests rather than creating advantages for him/her. EU law functions, therefore, fully in malam partem. Should this be the scenario, national judges, in engaging with EU provisions deprived of direct effect, may refer to the CJ in case of doubts regarding the scope or the internal effect of EU law, or, alternatively, must refer the matter to national constitutional courts, provided that such mechanism is envisaged under the domestic legal system, as it is in Italy. As a matter of fact, given the impossibility of relying on the remedy of consistent interpretation, in both Lucchini and Taricco, the Italian judges should have referred the matter to the Constitutional Court, rather than rely on the disapplication of domestic norms. This should have been done not because, as emerged in the Taricco saga, a counterlimit was at stake, but because EU law was not directly effective.

7. The future of direct effect and its effet utile

In the previous sections, I claimed that the scope of direct effect goes beyond the Van Gend & Loos ruling and the doctrine deriving from it. The principal arguments are summarised below.

Firstly, direct effect has two facets since, aside from a subjective-substitutive form of direct effect, there exists an objective-oppositive manifestation of direct effect, whereby a directly effective EU provision entails the disapplication of national law, without either conferring immediately a right upon the individual nor replacing the domestic norm in governing the case at hand. Secondly, the obligation to disapply can be triggered only by both primacy and direct effect, never by primacy alone. Thirdly, disapplication, while being normally the immediate, unavoidable result of direct effect, may not come into play if the CJEU deems that it could unduly harm the individuals and/or if national identity clause could be successfully invoked by the concerned Member State. Fourth, the classic test for assessing whether an EU norm is directly effective is no longer a pillar in the conceptualisation and practice of direct effect. Fifth, direct effect and direct applicability, albeit theoretically distinct, are de facto equivalent concepts since, in practice, unconditionality, on the one hand, coincides with direct applicability and, on the other, proves to be the only core element of the test of direct effect. Sixth, there is a further distinctive feature of direct effect, along with the unconditional character of an EU provision: the creation of an advantage, a benefit that the individual derives from the application of EU law and the subsequent disapplication of national law. Unconditionality/direct applicability and the creation of an advantage − through subjective-substitutive direct effect or objective-oppositive direct effect − are thus the only two true cumulative conditions for direct effect. Therefore, direct effect cannot arise when the application of the EU rules is only in malam partem, ie, when EU law only harms the individuals who are parties to the dispute. Thus, even when an EU provision is directly applicable/unconditional (first condition), in so far as the existence of an advantage for the individual (second condition) is lacking, direct effect will not occur and thus disapplication shall not occur.

Against this background, reconstructing and revisiting direct effect is crucial for (re)affirming the relevance of a foundational doctrine of EU law. Direct effect is a nexus, a synthesis between different legal traditions. It is a symptom and instrument of integration, a compass that helps the interpreters of the law to find their way among an indefinite, intricate cabala of varying categories of substantive and procedural law. Direct effect continues to be a fundamental principle, an essential doctrine capable, in its interaction with primacy, of shaping in proto-federalist terms a legal order that, albeit juridically mature and unique in its kind, is not a federal union. ²⁰⁵ Should the EU ever become a federal union, direct effect, as a filter and sine qua non condition for ensuring the effective protection of individuals, will no longer be necessary. Only then will national courts have to handle EU law in the same way they handle domestic law and, therefore, apply EU provisions, regardless of their nature, as directly effective norms.

Direct effect is not to be deemed as a (short-lived) ‘infant disease of EU law’ or something that, originally conceived as the normal state of health of the law, was supposed to spur the process of integration and then quietly disappear. ²⁰⁶ Direct effect is not a ‘useless’ or ‘obsolete’ concept’. ²⁰⁷

Nevertheless, the doctrine of direct effect must be refined and clarified. This duty should be carried out by the CJEU with a view to providing a guide for national authorities. I do not think that, as of today, the EU case law, often too apodictic and contradictory when it comes to the content, scope, boundaries, and consequences of direct effect, is capable of assuring sufficient certainty of the law. And this is serious since, when EU primi inter pares law constructs, such as direct effect and primacy, are at stake, the CJEU should use its best efforts to understand legal institutions and establish the applicable legal framework so as to prevent the risk of inconsistencies amongst judiciaries across the EU and, ultimately, avoid discrimination amongst individuals and companies due to national self-centred reconstructions of direct effect.

Now, understatement for a judge is normally certainly a merit; nonetheless, if the judge is supranational and has to guide 27 different jurisdictions, argumentative minimalism brought to its extreme and fundamentalist self-restraint, on the part of EU judges, when direct effect applies, undermines the spirit implied in the process of European integration. This holds true especially in a time like ours, when new forms − even court-based − of populism, sovereignism, and anti-Europeanism are on the rise. ²⁰⁸

Only the CJEU can offer guidance to national authorities, since it has an interpretive monopoly on the ifs, whens, and hows of direct effect. It is vital that the EU judges come to reassert their constitutional role and establish the common core of the EU system, beginning with the principles that created and shaped it. To counter the departure of national courts from EU constitutional law, as interpreted by EU judges, and try to mitigate the judicial activism and judicial rebellion ²⁰⁹ of national adjudicating bodies as much as possible, the CJEU should go back to doing what it boldly did over a long period of time during the sixties and seventies. In other words, it should ‘take seriously’ its role as a key actor in the systematisation of the principle of direct effect, in its engagement with the principles of primacy and effectiveness/effet utile. ²¹⁰

Indeed, the judicial recognition of direct effect through the avenue of the preliminary reference procedure – rather than its establishment at the primary law level –, far from being a limit per se, allows the doctrine to mature and evolve. The strength of principles deriving from the EU case law, as is the case for direct effect, lies in the CJEU’s ability to adjust them and, if necessary, modify their content while, at the same time, ensuring a steady dialogue with the national courts.

In conclusion, in order for an integrated and superordinate system of rights to be strengthened, it is indispensable that the court which established that system, ²¹¹ rather than simply acting as a ‘bouche qui prononce les paroles de la loi’, ²¹² perform, for real, a policy making function. In this light, clarifying and rethinking direct effect – which means ensuring legal certainty and consistency in the interpretation and application of EU law, as well as fostering the individual guarantees connected with it – is one of the greatest challenges that the CJEU has ever faced.