Hostname: page-component-77c89778f8-m8s7h Total loading time: 0 Render date: 2024-07-20T23:15:11.272Z Has data issue: false hasContentIssue false

Is the Primacy of EU Law Based on the Equality of the Member States? A Comment on the CJEU’s Press Release Following the PSPP Judgment

Published online by Cambridge University Press:  28 August 2020


In response to the BVerfG’s PSPP judgment, the CJEU issued an unprecedented press release in which it claimed that national courts must give full effect to EU law because this is the only way to ensure the equality of the Member States. While the press release does not contain the word “primacy,” the obligation to give full effect to EU law clearly implies primacy of EU law. This article provides a constructive and critical analysis of the press release’s “equality argument.” First, it demonstrates how the CJEU most likely borrowed it from a recent article by Federico Fabbrini, and how the argument is virtually identical to Kelsen’s defense of international law primacy. Second, it criticizes the equality argument for being inconsistent with the CJEU’s case law, and shows how it is either wrong or tautological. After suggesting three possible reasons why the press release nonetheless justifies primacy in this way, the article concludes by showing how the CJEU’s case law is better conceptualized from a Hamiltonian perspective in which full effect is a goal rather than a means.

Creative Commons
Creative Common License - CCCreative Common License - BY
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (, which permits unrestricted re-use, distribution, and reproduction in any medium, provided the original work is properly cited.
© The Author(s), 2020. Published by Cambridge University Press on behalf of the German Law Journal

A. Introduction

In response to the PSPP judgment of the Bundesverfassungsgericht (BVerfG),Footnote 1 the Court of Justice of the European Union (CJEU) issued an unprecedented press release, following the “many enquiries” which it had received. After noting that “[t]he departments of the institution never comment on a judgment of a national court,” the press release continues as follows:

“In general, it is recalled that the Court of Justice has consistently held that a judgment in which the Court gives a preliminary ruling is binding on the national court for the purposes of the decision to be given in the main proceedings.Footnote 2 In order to ensure that EU law is applied uniformly, the Court of Justice alone—which was created for that purpose by the Member States—has jurisdiction to rule that an act of an EU institution is contrary to EU law. Divergences between courts of the Member States as to the validity of such acts would indeed be liable to place in jeopardy the unity of the EU legal order and to detract from legal certainty.Footnote 3 Like other authorities of the Member States, national courts are required to ensure that EU law takes full effect.Footnote 4 That is the only way of ensuring the equality of Member States in the Union they created.”Footnote 5

One may question whether a press release is the appropriate genre for the CJEU to engage with its interlocutors, especially as selectivity is an easy charge: there was no such response to the Czech Constitutional Court’s decision in Slovak Pensions,Footnote 6 nor to that of the Danish Supreme Court in Ajos.Footnote 7 This article will leave questions of decorum aside, however, instead focusing on the substance of the CJEU’s message.

Remarkably, the Court managed to succinctly summarize its well-known case law on the foundational doctrines of EU law without mentioning any of the words “autonomy,” “direct effect,” “supremacy,” or “primacy”. This semantics is undoubtedly deliberate.Footnote 8 In substance, the first three sentences of the press release contain no surprises for EU lawyers. In the last two sentences, however, the Court makes a puzzling claim: national courts must give full effect to EU law because this is the only way to ensure the equality of the Member States. The statement is puzzling for two reasons. It is a justification for the primacy of EU law foreign to the Court’s own jurisprudence, and at least to me it is entirely unclear what it means.

This article aims to provide a constructive and critical engagement with the press release’s claim that primacy is the only way to ensure the equality of the Member States (hereinafter, “the equality argument”). To this end, section B seeks to first understand what the claim exactly means. Absent clarification in the press release itself, I will use two other sources as interpretive guidance: an interview with CJEU President Koen Lenaerts after the BVerfG’s judgment,Footnote 9 and a recent article by Federico Fabbrini in which he puts forward the equality of the Member States as a new rationale for the supremacy—or primacyFootnote 10—of EU law.Footnote 11 Given that the press release and Lenaerts’ elaboration are almost identical to Fabbrini’s new argument for absolute primacy, I gather that the latter has been the Court’s major source of inspiration. Because the equality argument alludes to Hans Kelsen’s work on the primacy of international law, I also discuss to what extent it can be understood in this manner.

Subsequently, section C offers a critique of the equality argument. My criticism is threefold: the logic of the equality argument mistakes enforcement for primacy; the equality argument cannot account for the “internal primacy” of EU law; and the full effect of EU law has always been understood by the CJEU’s case law as a goal, not a means. Following this criticism, I suggest three possible reasons why the press release justifies primacy in this way (section D). Finally, I briefly show how the CJEU’s case law is nonetheless better conceptualized from a Hamiltonian perspective, as opposed to the Kelsenian tone of the press release (section E).

B. The Argument Based on the Equality of the Member States

I. Primacy in a Multilateral Context

While the press release does not use the word “primacy”, the obligation to ensure the full effect of EU law clearly entails the application of EU law primacy as a basic conflict rule.Footnote 12 The reason why primacy is necessary to ensure the equality of the Member States is not explained by the press release. In an interview with the Dutch newspaper NRC shortly after the BVerfG’s judgment, however, President Lenaerts elaborates briefly on this point. In response to the question why EU law must have primacy, he answered:

“Because that primacy protects the equality of the Member States. If a dispute arises in which one party thinks: ‘I had understood this differently’, for instance about the ECB, then you go to the European Court because the Court delivers judgments for all contracting parties […] The Treaty stipulates that the Member States are equal. That is why the European Court is competent to provide a ‘uniform interpretation’ of Union law in all Member States. If one Member State does not comply with European judgments, other Member States can never use that as an excuse to also not comply with the rules. If it were otherwise, the first Member State ignoring a judgment would be able to unravel the entire European legal order.”Footnote 13

The answer is a mixture of norm-theoretical considerations (the primacy of norms of EU law over norms of national law) and institutional ones (the CJEU’s institutional prerogative to provide a uniform interpretation of EU law). Lenaerts’ first point centers not really on primacy as a norm-theoretical conflict rule, but rather on the CJEU’s exclusive competence to decide on the definitive meaning and validity of EU law. This aligns with the fact that the PSPP judgment did not deny the primacy of EU law as such, but rather the CJEU’s exclusive competence to decide on how EU law should be interpreted.Footnote 14 For Lenaerts, the combination of EU law primacy and the CJEU’s authority “to say what EU law is”Footnote 15 protects the equality of the Member States because it ensures a uniform interpretation of EU law that is valid for all Member States. This raises the question whether “the equality of the Member States” can be distinguished from “equal application of EU law in all Member States”—a point to which I will return in section C.

The conflation of primacy and uniformity is also central to Lenaerts’ second point. Absent primacy and the CJEU’s final say, one Member State’s failure to follow EU law could result in an unravelling of the EU legal order because of other Member States’ subsequent decisions to also ignore EU law.

Interestingly, while he first focuses on the value of a uniform interpretation of EU law, Lenaerts connects this to a seemingly different point. The straightforward argument for uniformity would be that diverging interpretations of EU law would threaten the unity of EU law, for instance because different national apex courts reach different conclusions on the delineation between “monetary” and “economic” policy.Footnote 16 Instead, Lenaerts mentions a different risk, namely that the non-compliance of Member State A would be a reason for Member State B to also not comply with (other parts of) EU law. This seems to point at the exception of non-performance (exceptio non adimpleti contractus) in international law, which allows States to withhold an obligation if another State has also withheld this or a similar obligation. This exception does not exist in EU law.Footnote 17

While I do not immediately see the link between this exception and EU law primacy, it is plausible that Lenaerts indeed had the exception of non-performance in mind. This is because his equality argument appears to be inspired by an article by Federico Fabbrini, which provides a new defense of EU law’s absolute primacy over conflicting national law.Footnote 18 Fabbrini precisely argues that primacy is necessary for the equality of Member States, and relies to this end on the multilateral context of primacy and the absence of an exception of non-performance.

According to Fabbrini, situations of competing primacy claims between EU law and national constitutional law should be evaluated in a multilateral perspective. Should a bilateral struggle between the CJEU and a national court be solved in favor of national law, or the national court’s interpretation of EU law (as in the PSPP judgment), this would inevitably affect all other Member States and their courts.Footnote 19 In the context of monetary policy, an eventual obligation for the Bundesbank to withdraw from the Public Sector Purchase Programme (PSPP) would have direct and significant consequences for the other central banks and the functioning and viability of monetary policy.Footnote 20 However, Fabbrini mainly seems to refer to the different risk that one Member State could get away with a failure to comply with EU law, thus casting doubt on the reciprocal nature of the Member States’ commitments.Footnote 21 He connects this multilateral perspective to the absence of the exception of non-performance in EU law. For Fabbrini, “[i]t is because the EU is supreme over national law—and the contracting parties abide by this rule—that the Member States are not entitled in the EU to tit-for-toe each other in case of violations of EU law.”Footnote 22

Thus, the absence of the exception of non-performance is justified by the primacy of EU law, and the argument runs as follows: firstly, primacy of EU law works in tandem with the CJEU’s competence to provide a uniform interpretation. Secondly, this uniform interpretation ensures that all Member States and their courts have an equal obligation to comply with EU law. Thirdly, for this reason there is no exception of non-performance. The comments by Lenaerts precisely echo this argument, which makes it reasonable to believe that the press release should be understood along these lines.Footnote 23

II. Has Primacy Turned Kelsenian?

Notwithstanding the self-standing relevance of the argument described above, one elephant in the room remains. This elephant is Hans Kelsen. While the doctrine of primacy in EU law is generally considered distinct from “ordinary” international law primacy,Footnote 24 the press release’s linkage between primacy and equality is also virtually a copy of Kelsen’s argument in favor of the primacy of international law. Kelsen believed that “the idea of the equality of all States can be maintained only if we base our interpretation of legal phenomena on the primacy of international law.”Footnote 25 This makes “international law monism”—which infers the validity of national law from international law—more coherent than “national law monism.”Footnote 26 The Westphalian ethic of the equality of states is only possible insofar as “above the communities understood as states stands a legal order that mutually delimits the spheres of validity of the individual states in that it hinders incursions by one into the sphere of the others, or at least subjects them all to equal conditions for such incursions.”Footnote 27

Does this mean that the primacy of EU law is nothing but the primacy of international law in a specific treaty context? This seems implausible because the CJEU’s case law has precisely denied that the EU legal order is merely a treaty under international law.Footnote 28 At the same time, it is not clear how the “equality of the Member States” is different from the “sovereign equality of States.”

In another article on the principle of mutual trust after Opinion 2/13,Footnote 29 Lenaerts offers some thoughts on this point. He argues that the equality of the Member States should be understood in light of the fact that all Member States are deemed to share the same degree of commitment to the values of the EU, enshrined in Article 2 TEU.Footnote 30 For Lenaerts, this is different in international law: the principle of sovereign equality “confers on every sovereign State an equal capacity to enter into international obligations but is silent as to whether States share a set of common values.”Footnote 31 Therefore, “[a]n EU Member State and a third country may be equals before international law, but they are not equals before the law of the EU as only the former is part of the EU, understood as a Union of values.”Footnote 32 I am not sure whether this argument can be extrapolated to parts of EU law which are less closely connected to Article 2 TEU than the principle of mutual trust in the context of the Area of Freedom, Security and Justice (AFSJ), but in which primacy is equally absolute (including the Economic and Monetary Union). It is of course true that the Member States are deemed to share a commitment towards the entire EU acquis and its binding nature, but this seems to simply reflect the formal difference between being a member and not being a member. Accordingly, I fail to see how this distinction could ground primacy in some value-based conception of equality of Member States as opposed to the sovereign equality of States.

However, there is a related distinction which I think is crucial for a proper understanding of the equality argument. On the one hand, there is the formal equality of Member States, which includes among others the fact that they are equally bound by the CJEU’s case law. On the other hand, the equal application of EU law within all Member States refers to a more, for lack of a better term, “substantive” or “effects-based” notion of equality. As I shall try to show in the next section, an “effects-based” understanding of the equality of the Member States is the only way in which the equality argument makes sense. However, this “effects-based” understanding makes it impossible to distinguish between full effect and equality, making the press release’s reasoning tautological.

C. Why the Equality Argument Is Unconvincing

I. The Equality Argument Mistakes Enforcement for Primacy

An important part of the equality argument is the alleged link between the role of primacy in EU law and the absence of an exception of non-performance. The latter is said not to exist because there is primacy. Unilateral non-performance would give the non-performing Member State a derogation from its EU law obligations, while all Member States should be equally bound by EU law.

However, the soundness of this reasoning remains obscure. The exception of non-performance exists in international law notwithstanding the fact that the latter also claims primacy over national law,Footnote 33 and this primacy arguably roots precisely in the ideal of sovereign equality. Furthermore, if equality is understood formally, an exception of non-performance seems to perfectly ensure the equality of the Member States because it would be available to all Member States equally, should its conditions apply. Surely, that would be disastrous for the uniform application of EU law. If that is the point, however, and the equality of the Member States simply refers to the equal and uniform application of EU law by and within all Member States, the distinction between “full effect” and “equality” becomes muddled. Article 4(2) TEU could no longer support the equality argument because the way in which the Member States are required to comply with EU law is immaterial to the Union respecting their equality, as long as these requirements apply equally.

The rationale of the exception of non-performance can rather be found in Kelsen’s analysis of the international legal order. Kelsen recognized the international legal order’s defects insofar as it has no adjudicatory institutions that can invalidate national laws breaching international law. Presaging HLA Hart’s account of the “primitive” nature of international law,Footnote 34 Kelsen pointed out that international law delegates the task of enforcement to the same entities which are its main subjects, i. e. States. International law is enforced not by way of judicial invalidation, but by sanctions including just war, reprisals and reciprocal non-performance.Footnote 35

The exception of non-performance is, accordingly, entirely irrelevant to both primacy and the equality of the Member States, and the correlation is wrong both ways. The doctrine exists in international law notwithstanding the latter’s primacy over national law. The reason why the doctrine does not exist in EU law is because the latter has a system of judicial protection which delegates the task of enforcement to national and EU courts.Footnote 36 The absence of an exception of non-performance is vital to the full effect of EU law. But to suggest that it has any relationship to EU law primacy is to mistake the latter for the modalities of enforcement which the legal order employs.

II. Internal Primacy Is Not Necessary for Equality

While the doctrine of international law’s primacy is necessary to ensure the Westphalian equality of states, this is neither how EU law primacy was introduced, nor how it is justified in subsequent jurisprudence. As opposed to international law, EU law characteristically also claims “internal primacy”.Footnote 37 This means that EU laws can have direct effect under criteria determined by EU law itself,Footnote 38 and consequently that they also have primacy within the national legal orders, legally obligating national courts to disapply all conflicting national norms.Footnote 39 For the CJEU, this doctrine is inferred from the EU Treaties’ creation of an independent legal order dissociated from ordinary international law. The equality argument should therefore be assessed not in light of ordinary international law primacy, but specifically in light EU law’s claim of “internal primacy”.

What is meant exactly by “equality” is, again, crucial. Insofar as Article 4(2) TEU obligates the EU to treat the Member States equally, internal primacy is surely not necessary to ensure such equal treatment. Suppose that the ECB fails to justify its PSPP and, consequently, the Bundesbank withdraws from the program. This might lead the European Commission to start infringement proceedings against Germany under Article 258 TFEU. In this scenario, the equality of the Member States is not compromised at all, except to the extent that the Commission has not started similar proceedings against, for instance, Denmark after the Danish Supreme Court’s Ajos judgment. This, however, points at the Commission’s exercise of discretion under Article 258 TFEU: it has nothing to do with the necessity of (internal) primacy of EU law.

The link between internal primacy and equality only makes sense if the latter refers to the equal, i. e. uniform, application of EU law within all national legal orders, for which internal primacy is indeed necessary. The press release should then be read as a response to the BVerfG’s claim that the CJEU’s judgment in Weiss “has no binding force in Germany.”Footnote 40 However, what else is the uniform application of EU law within all Member States but exactly the full effect of EU law? The highest degree of effectiveness of EU law surely is reached when all of EU law is applied uniformly by all courts in all Member States. Should the CJEU’s judgment in Weiss be valid law in all Member States but Germany, it is clear that the judgment would only have a partial effect.

Hence, the equivocation of “equality of the Member States” and “equal application of EU law within the Member States” makes the press release’s reasoning tautological. It says that all national courts must ensure the full effect of EU law because this is the only way to ensure that EU law is equally applied in all Member States, which is precisely what full effect means.

An argument which we may consider—even though it is not explicitly raised by Fabbrini or Lenaerts—is that uniform internal primacy is necessary to ensure the equality of Member States as soon as one Member State recognizes internal primacy (as the Netherlands did at the time of Van Gend & Loos). After establishing direct effect in virtue of EU law itself, internal primacy in all Member States is the only way to avoid inequalities in the way in which EU law is applied within the various Member States. Unfortunately, this brings back the distinction between “equality” and “full effect.” The equality of the Member States as such could be perfectly ensured by consistent infringement proceedings against all Member States in which the primacy of EU law would not be recognized by a national court of last instance, notwithstanding the fact this would be less effective than uniform internal primacy.

This conclusion connects to Fabbrini’s and Lenaerts’ reliance on the risk of a proliferation of “selective exits,”Footnote 41 which would be disastrous for the integrity and unity of EU law. Again, however, any means of consistent and equal enforcement against violations of EU law would respect the equality of the Member States equally well. The choice between centralized enforcement via Articles 258–260 TFEU (and Article 7 TEU), and decentralized enforcement within the Member States (presuming internal primacy) is irrelevant for ensuring the equality of the Member States except to the extent that the latter entirely collapses into full effect.

Perhaps the press release’s only point is that the BVerfG should not be allowed to claim prerogatives that less famous apex courts lack as well. Read this way, however, the reasoning of the press release should be exactly the opposite: equal treatment of all Member States and their apex courts is necessary to ensure full effect, not the other way around.

III. Full Effect Is a Goal, Not a Means

The previous two sections aimed to question the logic of the claim that full effect and primacy are necessary to ensure the equality of the Member States. Perhaps most remarkable about the re-engineering of primacy, however, is that it conceives of full effect as merely a means in the first place. This claim is deeply incoherent with the CJEU’s own jurisprudence, which has consistently conceived of the effectiveness of EU law as a goal in itself, not as a means.

Already in Costa v. ENEL, the relationship between full effect and equality seems the opposite of how it is portrayed in the CJEU’s press release: “The executive force of Community law cannot vary from one State to another in deference to subsequent domestic laws, without jeopardizing the attainment of the objectives of the Treaty […]”Footnote 42 In other words, the equal application of EU law in all Member States is necessary to ensure the effective attainment of the Treaty’s objectives.

This conceptualization of primacy as a doctrine that aims to guarantee the full effect of EU law as an objective in itself is consistent with the self-standing role of effectiveness in other contexts, notably direct effect. In Van Gend & Loos, the doctrine of direct effect was considered necessary because it would enable “an effective supervision in addition to the supervision entrusted by Articles [258 and 259 TFEU]” thanks to “the vigilance of individuals concerned to protect their rights.”Footnote 43 The gradual loosening of the criteria for direct effect,Footnote 44 combined with an expansion of its applicability to decisions, directives, and general principles, also explicitly referred to full effect as a self-standing goal.Footnote 45

While some consider this problematic from a hermeneutic perspective,Footnote 46 aspiring towards full effectiveness is understandable for a new legal order that is yet to prove its sustained viability. Pescatore’s analysis of the relevance of effectiveness to EU law remains illuminating:Footnote 47

“[a]ny legal rule is devised so as to operate effectively (we are accustomed, in French, to speak here about effet utile). If it is not operative, it is not a rule of law […] In other words, practical operation for all concerned, which is nothing else than ‘direct effect’, must be considered as being the normal condition of any rule of law.”Footnote 48

Since direct effect without internal primacy is manifestly ineffective, it is no surprise that internal primacy was considered equally necessary for full effectiveness in the early 1960s.Footnote 49

The Court’s preoccupation with effectiveness as a goal—especially in relation to primacy—also took center stage in its decisions in Opinion 2/13 and Achmea. The effectiveness of the preliminary reference procedure and in particular the CJEU’s prerogative to decide on the proper interpretation of EU law have one clear objective: “securing uniform interpretation of EU law, thereby serving to ensure its consistency, its full effect and its autonomy as well as, ultimately, the particular nature of the law established by the Treaties.”Footnote 50 Nothing in the reasoning of the Court’s case law suggests that these “mechanisms capable of ensuring the full effectiveness of the rules of the EU”Footnote 51 serve any further goal than full effectiveness of the EU legal order.

In other words, nothing in the jurisprudence suggests that equality has ever been a consideration in requiring EU law primacy, except insofar as “equality of the Member States” actually means “equal application of EU law in the Member States.” However, this effects-based understanding of equality dismantles Lenaerts’ argument about the textual basis of primacy,Footnote 52 as Article 4(2) TEU merely requires the EU to treat Member States equally and is agnostic about the way in which the EU legal order deals with conflicting national laws. More importantly, it turns the press release’s reasoning into a tautology. The good news is that the press release is not really at odds with the case law. It just re-phrases the case law in an extremely awkward way, by substituting “equal effect within the Member States” for “equality of the Member States” and then distinguishing the latter from “full effect” while in reality we are speaking about the same thing. This might be a rhetorically helpful way of putting things, as I suggest in the following section. But that does make it neither coherent nor right.

D. The Attractiveness of the Equality Argument

If I am right in rejecting the link between primacy and the equality of the Member States, this raises the question what were reasons to formulate the press release in this manner. In this section, I suggest three reasons that may underlie this formulation.

I. Backlash Against Autonomy and Absolute Primacy

As mentioned above, the press release tries to re-engineer EU law primacy without mentioning any of the words “primacy,” “supremacy,” or “autonomy.” This has, no doubt, been a deliberate choice. If the doctrines of autonomy and absolute primacy are understood as a change in the Grundnormen of the national legal orders,Footnote 53 it goes without saying that the Court’s project has failed. National constitutional courts overwhelmingly do not recognize the supremacy of EU law over their national constitutions. In Davies’ words, the CJEU “played, and it lost.”Footnote 54

While this is nothing new,Footnote 55 there is much reason to believe that the current situation is more critical than usual. The PSPP judgment is the first truly consequential attack, by one of Europe’s most important apex courts, on the autonomy and absolute primacy of the EU legal order and the CJEU’s final say. A reiteration of doctrines whose existence is precisely denied by the BVerfG indeed seems hardly useful.

Secondly, the PSPP judgment should be seen against the background of the multiple blows that primacy and autonomy had already received in recent years. One may point at the CJEU’s initial interpretation of the requirements of mutual trust in AFSJ matters,Footnote 56 which proved incompatible with the jurisprudence of the European Court of Human Rights,Footnote 57 and which the CJEU’s accordingly changed in a face-saving re-interpretation of its case law.Footnote 58 In turn, the Court’s preoccupations with autonomy to the detriment of EU accession to the ECHR in Opinion 2/13 were almost universally criticized,Footnote 59 and scholarly views on its Achmea judgment on the incompatibility of intra-EU BIT agreements with the autonomy of EU law have been equally critical.Footnote 60

Following Achmea, Opinion 1/17’s more lenient approach to autonomy, generally unexpected as it was,Footnote 61 may be a sign that the Court is responsive to this critique.Footnote 62 Against this background, the press release may be understood as further confirmation that the CJEU has understood the message: absolute primacy and autonomy are controversial words, better to be avoided.

II. Appealing to International Law and the Treaty Text

The previous point connects to two obvious points in favor of the equality argument. The first one is that it alludes to the sovereign equality of States and is identical to the Kelsenian justification of international law primacy. Its logical weaknesses in the EU context notwithstanding, at face value the argument may appear powerful. The language of international law is less rebellious than the CJEU’s traditional parlance of autonomy and absolute primacy. It appeals to a vocabulary that all Member States share and accept.

Secondly, the equality argument appeals to the text of the Treaty. In the past, many scholars have questioned the absolute primacy of EU law because the Treaties lack a supremacy clause like the US Constitution.Footnote 63 Prior to the inclusion of the EU’s obligation to ensure the equality of the Member States in Article 4(2) TEU, relying on the equality of the Member States would not have had the rhetorical advantage of giving primacy a textual basis.Footnote 64 It may be somewhat ironic that the rationale of primacy is re-engineered in terms of the provision that some had thought would be the end of absolute primacy.Footnote 65 In the current constitutional landscape, however, Article 4(2) is a rhetorically powerful reference.

III. “Selective Exit” as a Common Nightmare

Finally, the substantive concerns about the possible implications of the PSPP judgment are well-founded. Reporters were quick to analyze the judgment as a “bombshell”Footnote 66 which could threaten monetary policy and the Euro.Footnote 67 More broadly, the BVerfG’s reasoning clearly provides a helpful template for other Member States to justify their own invalidations of some other part of EU (case) law.

This article is not to deny in any way, therefore, the serious risks of an unraveling of the EU legal order through a series of “selective exits” by Member States. No legal system exists that is not generally obeyed.Footnote 68 Selective non-performance of legal obligations has been casting doubt on the legality of international law for many years.Footnote 69 Beyond monetary policy as such, the PSPP judgment may pose a serious threat to the full effect of EU law, and to the extent that this is the real message of the press release, it seems entirely warranted. Its value notwithstanding, “full effect” is no real legal term of art because all legal norms aspire towards having full effect. The equality of the Member states—firmly established in Article 4(2) TEU—surrounds concerns about effectiveness with an aura of juridical reasoning proper—albeit thin.

E. Primacy of EU Law: Hamiltonian, Not Kelsenian

The question remains what is the basis of absolute primacy of EU law, if any, and whether the CJEU could have done a better job in conveying its importance. While this is not the place to elaborate extensively on this question, briefly juxtaposing the press release’s quasi-Kelsenian perspective with what is in my view a more coherent rationale seems in place.

EU law primacy is best understood from the perspective of the plurality of legal orders, each of which claims primacy over competing legal and non-legal normative systems.Footnote 70 Accordingly, primacy should first and foremost be considered a doctrine internal to EU law, reflecting the EU legal order’s desire to become and remain effective. This is the only way in which it can be taken seriously as “law.” How central primacy is to the CJEU’s conception of legal order becomes clear when the famous dicta in Costa v. ENEL about the conceptual connection between the legality of the Rome Treaty and the primacy of EU lawFootnote 71 are compared to Alexander Hamilton’s elaborations on the US Constitution’s supremacy clause, which he considered merely declaratory:

“But it is said [in the US Constitution, Article 6, § II] that the laws of the Union are to be the supreme law of the land. But what inference can be drawn from this, or what would they amount to, if they were not to be supreme? It is evident they would amount to nothing. A LAW, by the very meaning of the term, includes supremacy. It is a rule which those to whom it is prescribed are bound to observe. […] If a number of political societies enter into a larger political society, the laws which the latter may enact, pursuant to the powers intrusted to it by its constitution, must necessarily be supreme over those societies, and the individuals of whom they are composed. It would otherwise be a mere treaty, dependent on the good faith of the parties, and not a government, which is only another word for POLITICAL POWER AND SUPREMACY.”Footnote 72

From this Hamiltonian viewpoint, I do not think there is a useful distinction between primacy and supremacy in the way Matej Avbelj among others has suggested.Footnote 73 When Hamilton speaks about supremacy, he makes the elementary claim that a law would not be a law if it did not claim supremacy. Similarly, the CJEU in Costa v. ENEL presents primacy as a necessary consequence of the fact that the EU Treaties have created an independent legal order whose norms are binding as law on its subjects. In other words, the primacy and binding nature of EU law are inevitable given the latter’s claim to be “law” properly speaking. There is no Westphalian ethic here that requires primacy in order to make sense of the sovereign equality of States. To the contrary: the equal application of EU law by and within all Member States is merely instrumental towards full effect.

Internal primacy of EU law, more specifically, is implied by the CJEU’s understanding of national courts being “decentralized EU courts,”Footnote 74 or in Lenaerts’ words, “arms of EU law.”Footnote 75 The obligation to ensure the full effect of any law, after all, extends to all those who are considered “arms of the law”, making the question of the identity of national courts logically prior to that of internal primacy. The manner in which the CJEU considers national courts arms of EU law is as Hamiltonian as its doctrine of primacy. It was Hamilton who introduced the idea that the legal officials of the US States would be auxiliary enforcers of the national government in the nascent US federal order:

“Thus the Legislatures, Courts and Magistrates of the respective members will be incorporated into the operations of the national government, as far as its just and constitutional authority extends; and will be rendered auxiliary to the enforcement of its laws.”Footnote 76

Hamilton’s perspective on the US federal order provides a blueprint for how the CJEU conceives of the EU legal order, and explains the rationale of primacy—including internal primacy—without postulating any further goals beyond any legal order’s desire to become and remain effective. This is fully in line with the CJEU’s own case law.

Obviously, none of this solves any of the problems associated with a clash of legal orders. I think this impractical conclusion has some value nonetheless, precisely in what it denies; that one can defend the primacy of EU law legally. The absolute primacy of EU law cannot be justified on legal grounds because it is a quality of the concept of “law” itself.

F. Conclusion

Among the many important questions surrounding the PSPP judgment, two sentences from a non-binding press release may not seem most relevant or exciting. Nonetheless, the press release too raises interesting questions, not only regarding the question of whether issuing it was strategically sound, but also regarding the provenance and the robustness of its substance. As I have tried to show, it is highly plausible that the CJEU has silently borrowed Fabbrini’s innovative defense of the absolute primacy of EU law, and used it to substantiate its own position in a vocabulary devoid of controversial terms.

Accordingly, I read the press release as a creative attempt by the CJEU to maintain its stance while realizing that the old arguments have proved unconvincing for some of its interlocutors, to strength I would not discuss the question whether issuing a press release was a sound choice, in one way it clearly is not: a succinct and weakly argued press release makes the latter vulnerable to obvious objections to its reasoning; not quite useful for a Court the quality of whose legal reasoning is currently contested,Footnote 77 and is questioned more generally.Footnote 78

As Hamilton saw intuitively—long before modern legal theory reached strikingly similar conclusionsFootnote 79—it is the word “law” itself that includes a primacy claim. To claim primacy means to claim that one ought to be effective. But what law, and what interpretation of the law, should be effective is left unanswered.Footnote 80 The persuasiveness of primacy and the CJEU’s final say is in the end a matter of ethics.Footnote 81 Article 4(2) TEU offers no help in this regard, nor does anything else legal.



Assistant Professor of European Law, University of Groningen; Emile Noël Fellow, NYU School of Law (2019–2020). I am thankful to Matthias Goldmann for useful comments on a previous draft. The usual disclaimer applies.


1 Bundesverfassungsgericht [BVerfG] [Federal Constitutional Court], Case No. 2 BvR 859/15 (May 5, 2020), [hereinafter Judgement of May 05, 2020].

3 ECJ, Case C-314/85, Foto-Frost, ECLI:EU:C:1987:452 (Oct. 22, 1987), paras. 15, 17.

5 ECJ Press Release 58/20, Press Release Following the Judgment of the German Constitutional Court of 5 May 2020 (May 8, 2020),

6 Czech Constitutional Court, Pl. ÚS 5/12, Slovak Pensions, Judgment of Jan. 31, 2012.

7 Supreme Court of Denmark, Case No. 15/2014, Judgment of Dec. 6, 2016.

8 See infra section D.

9 Caroline de Gruyter, President Koen Lenaerts: “Europese Hof komt meer center stage”, NRC (May 17, 2020),

10 Some authors claim there is an important distinction between these terms. However, I consider “primacy” and “supremacy” synonymous, for reasons briefly explained in section E infra.

11 Federico Fabbrini, After the OMT Case: The Supremacy of EU Law as the Guarantee of the Equality of the Member States, 16 German L.J. 1003 (2015).

12 ECJ, Case 106/77, Amministrazione delle Finanze dello Stato v. Simmenthal SpA, ECLI:EU:C:1978:49 (Mar. 9, 1978), paras. 17–24,; ECJ, Case 103/88, Fratelli Costanzo v. Comune di Milano, ECLI:EU:C:1989:256 (June 22, 1989), para. 33,; ECJ, Case C-213/89, The Queen v. Factortame, ECLI:EU:C:1990:257 (June 19, 1990), paras. 21–23,

13 de Gruyter, supra note 9 (my translation).

14 Judgment of May 5, 2020 at paras. 112–113.

15 See generally Koen Lenaerts & José A. Gutiérrez-Fons, To Say What the Law of the EU Is: Methods of Interpretation and the European Court of Justice, 20 Columbia J. Eur. L. 3 (2014).

16 Compare e. g. Judgment of May 5, 2020 at paras. 133–138, with ECJ, Case C-493/17 Heinrich Weiss and Others, ECLI:EU:C:2018:1000 (Dec. 11, 2017), paras. 53–70,

17 Joined Cases 90 & 91/63, Comm’n v. Luxembourg & Belgium, ECLI:EU:C:1964:80 (Nov. 13, 1964),

18 Fabbrini, supra note 11, at 1014.

19 Fabbrini, supra note 11, at 1014–1015.

20 Francesco Canepa & Balazs Koranyi, ECB Prepares for the Worst: Life Without the Bundesbank, Reuters (May 26, 2020),

21 Fabbrini, supra note 11, at 1016. I say “seems” because to me the point is not entirely clear.

22 Fabbrini, supra note 11, at 1016.

23 I assume that the press release has been drafted by, or in close cooperation with, the President’s cabinet.

24 Bruno de Witte, Direct Effect, Primacy and the Nature of the Legal Order, in The Evolution of EU Law 323, 324–326 (Gráinne de Búrca & Paul Craig eds., 2011). But see id., at 361–362.

25 Hans Kelsen, General Theory of Law and State 387 (2006 ed., 1945).

26 David Dyzenhaus, Kelsen’s Contribution to Contemporary Philosophy of International Law 29–31 (Apr. 8, 2020), at

27 Hans Kelsen, Das Problem der Souveränität und die Theorie des Völkerrechts: Beitrag zu Einer Reinen Rechtslehre 205 (1981, reprint of 2nd. edn., 1928) (cited in id., at 27).

28 See generally Case 6/64, Costa v. E.N.E.L., 1964 E.C.R. 585 (1964).; ECJ, Joined Cases C-402/05 P and C-415/05 P, Kadi and Al Barakaat, ECLI:EU:C:2008:461 (Sept. 3, 2008),

29 Koen Lenaerts, La vie après l’avis: Exploring the Principle of Mutual (Yet Not Blind) Trust, 54 Common Mkt. L. Rev. 805 (2017).

30 Id. at 808–809.

31 Id. at 809.

32 Id.

33 Treatment of Polish Nationals and Other Persons of Polish Origin or Speech in Danzig Territory, Advisory Opinion, 1932 P.C.I.J. (ser. A/B) No. 44, 24 (Feb. 4, 1932); Vienna Convention on the Law of Treaties, art. 27, May 23, 1969, 1155 U.N.T.S. I-18232.

34 H.L.A. Hart, The Concept of Law 91–94, 227–237 (3rd. edn., 2012).

35 Kelsen, supra note 25, at 338–341. Irrespective of whether the exception of non-performance is considered exactly analogous to a “sanction” (e.g. Jan Klabbers , International Law 167 (2013)), the Kelsenian argument about “responsive” sanctions applies equally to it. See generally Maria Xiouri, The Exceptio Non Adimpleti Contractus in Public International Law, 21 Int’l Comm. L. Rev. 56 (2019) (conceiving of the exception as a principle that fill gaps in the regulation of breaches of treaties).

36 Comm’n v. Luxembourg & Belgium, Joined Cases 90 & 91/63 at 631 (“[The Treaty] establishes a new legal order which governs the powers, rights and obligations of the said persons, as well as the necessary procedures for taking cognizance of and penalizing any breach of it. Therefore, except where otherwise expressly provided, the basic concept of the Treaty requires that the Member States shall not take the law into their own hands.”) (italics added).

37 de Witte, supra note 24, at 324–329.

38 ECJ, Case C-26/62, Van Gend en Loos v. Administratie der Belastingenat, ECLI:EU:C:1963:1 (Feb. 5, 1963), 12–13,

39 Costa v. ENEL, Case 6/64 at 593–594; Simmenthal, Case 106/77 at para. 21.

40 Judgment of May 5, 2020 at para. 163.

41 The term “selective exit” is used by J.H.H. Weiler, The Transformation of Europe, 100 Yale L.J. 2403, 2412 (1991), and points at selective non-performance of EU law obligations by a Member State.

42 Costa v. ENEL, Case 6/64 at 594.

43 Van Gend & Loos, Case 26/62 at 13.

44 Robert Schütze, Direct Effects and Indirect Effects, in Oxford Principles of European Union Law 268–270 (Robert Schütze & Takis Tridimas eds., 2018); Lorenzo Squintani & Justin Lindeboom, The Normative Impact of Directives: Casting Light on Direct Effect and the Elusive Distinction between Obligations and Mere Adverse Repercussions, 38 YB. Eur. L. 18, 22–24 (2019).

45 ECJ, Case Case 9/70, Grad v. Traunstein, ECLI:EU:C:1970:78 (Oct. 6, 1970), para. 5,; ECJ, Case 41/74, Van Duyn v. Home Office, ECLI:EU:C:1974:133 (Dec. 4, 1974), para. 12,; ECJ, Case C-144/04, Mangold v. Helm, ECLI:EU:C:2005:709 (Nov. 22, 2005), para. 77, See also Sibylle Seyr, Der effet utile in der Rechtsprechung des EuGH 122–33 (2008).

46 See generally Gerard Conway, The Limits of Legal Reasoning and the European Court of Justice (2014); Hjalte Rasmussen, On Law and Policy in the European Court of Justice (1986).

47 As Pescatore refers to “effet utile” it is important to note that “full effect” seems to denote a higher degree of effectiveness than “useful effect.” The case law, however, appears to use them alternatingly and without clear conceptual difference. See e.g. Hans Kutscher, Methods of Interpretation as Seen by a Judge at the Court of Justice, in Reports of a Judicial and Academic Conference held in Luxemburg 41–42 (1976); Seyr, supra note 45, at 111–114, 367.

48 Pierre Pescatore, The Doctrine of “Direct Effect”: An Infant Disease of Community Law, 8 Eur. L. Rev. 155, 155 (1983).

49 Amedeo Arena, From an Unpaid Electricity Bill to the Primacy of EU Law: Gian Galeazzo Stendardi and the Making of Costa v. ENEL, 30 Eur. J. Int’l L. 1017, 1026–1031 (2019); Morten Rasmussen, Revolutionizing European Law: A History of the Van Gend en Loos Judgment, 12 Int’l J. Const. L. 136, 154–156 (2014).

50 ECJ, Case Opinion 2/13, ECLI:EU:C:2014:2454 (Dec. 18, 2014), para. 176,; ECJ, Case C-284/16, Slowakische Republik v. Achmea, ECLI:EU:C:2018:158 (Mar. 6, 2018), para. 37,

51 Achmea, Case C-284/16 at para. 43.

52 See text accompanying note 13 supra.

53 J.H.H. Weiler, The Community System: The Dual Character of Supranationalism, 1 YB. Eur. L. 267, 275 (1981). In my view, this is not the best conceptualization of the primacy jurisprudence, as I argue elsewhere: Justin Lindeboom, The Autonomy of EU Law: A Hartian View, 12 Eur. J. Leg. Stud. (2020, forthcoming).

54 Gareth Davies, Interpretative Pluralism within EU Law, in Research Handbook on Legal Pluralism and EU Law 323, 323 (Gareth Davies & Matej Avbelj eds., 2018).

55 See generally Karen Alter, Establishing the Supremacy of European Law (2001); Bill Davies, Resisting the ECJ: Germany’s Confrontation with European Law, 1949–1979 (2012).

56 See generally ECJ, Joined Cases C-411 & 493/10, N.S., ECLI:EU:C:2011:865 (Dec. 21, 2011),

57 See generally M.S.S. v. Belgium & Greece 2011 Eur. Ct. H. R. 108,; Tarakhel v. Switzerland 2014 Eur. Ct. H. R. 1185,

58 ECJ, Case C-578/16 PPU, C.K. and Others v. Republika Slovenije, ECLI:EU:C:2017:127 (Feb. 16, 2017), para. 65,

59 See generally Eleanor Spaventa, A Very Fearful Court? The Protection of Fundamental Rights in the European Union after Opinion 2/13, 22 Maastricht J. Eur. Comp. L. 35 (2015); Piet Eeckhout, Opinion 2/13 on EU accession to the ECHR and Judicial Dialogue—Autonomy or Autarky?, 38 Fordham Int’l L.J. 955 (2015); Bruno de Witte & Šejla Imamovic, Opinion 2/13 on Accession to the ECHR: Defending the EU Legal Order Against a Foreign Human Rights Court, 40 Eur. L. Rev. 683 (2015).

60 See generally Csongor István Nagy, Intra-EU Bilateral Investment Treaties and EU Law After Achmea, 19 German L.J. 981 (2018); Panos Koutrakos, The Autonomy of EU Law and International Investment Arbitration, 88 Nordic J. Int’l L. 41 (2019).

61 E.g. Mauro Gatti, Opinion 1/17 in Light of Achmea: Chronicle of an Opinion Foretold, 4 Eur. Papers 109, 120 (2019).

62 See generally Christian Riffel, The CETA Opinion of the European Court of Justice and its Implications—Not that Selfish After All, 22 J. Int’l Econ. L. 503.

63 See e. g. Gerard Conway, The Limits of Legal Reasoning and the European Court of Justice 29–30 (2014); Thomas Horsley, The Court of Justice of the European Union as an Institutional Actor: Judicial Lawmaking and its Limits 115–131 (2018).

64 In Art. 4(2) TEU’s predecessors—Art. 6(3) TEU (Amsterdam) and Art. F(1) TEU (Maastricht)—respect for the equality of the Member States was not included.

65 Armin von Bogdandy & Stephan Schill, Overcoming Absolute Primacy: Respect for National Identity under the Lisbon Treaty, 48 Common Mkt. L. Rev. 1417 (2011).

66 Phillip Inman, Jolt to Eurozone as German Court Warns Against Central Bank Stimulus, The Guardian (May 5, 2020),

67 Ben Hall & Martin Arnold, German Court Ruling Casts Doubt on European Monetary Policy, Financial Times (May 6, 2020),

68 This is a truism of legal theory, see e. g. Kelsen, supra note 25, 119; Hart, supra note 34, 116–117.

69 But see Alexander Somek, Kelsen Lives, 18 Eur. J. Int’l L. 409 (2007).

70 Justin Lindeboom, Why EU Law Claims Supremacy, 38 Oxf. J. Leg. Stud. 328, 335–338 (2018). See generally, Joseph Raz, Practical Reason and Norms 151–154 (1975). In terms of the literature on EU constitutional pluralism, my viewpoint is probably most similar to what Neil MacCormick calls “radical pluralism.” See generally Neil MacCormick, Risking Constitutional Collision in Europe?, 18 Oxf. J. Leg. Stud. 517 (1998).

71 Costa v. ENEL, Case 6/64 at 594 (“The law stemming from the Treaty, an independent source of law, could not, because of its special and original nature, be overridden by domestic legal provisions, however framed, without being deprived of its character as Community law.”).

72 The Federalist No. 33 (Alexander Hamilton) (italics and capitalization in original).

73 Cf. Matej Avbelj, Supremacy or Primacy of EU Law—(Why) Does it Matter?, 17 Eur. L.J. 744, 745–754 (2011). For Avbelj, EU law primacy is a “trans-systemic principle, which regulates the relationship between the autonomous legal orders,” (id., at 750) in contrast to supremacy, which he considers an intra-systemic feature inappropriate to the question of conflicts between legal orders.

74 Simmenthal, Case 106/77 at para. 21. The logic of the CJEU’s case law is conceptualized in more detail in Lindeboom, supra note 53.

75 Koen Lenaerts, Upholding the Rule of Law Through Judicial Dialogue, 38 YB. Eur. L. 1, 4 (2019).

76 The Federalist No. 27 (Alexander Hamilton) (italics in original).

77 Judgment of May 5, 2020 at paras. 116–118.

78 See generally J.H.H. Weiler, Judging the Judges—Apology and Critique, in Judging Europe’s Judges 235 (Maurice Adams et al. eds., 2013).

79 Joseph Raz, The Authority of Law ch. 2 (1979); Raz, supra note 70.

80 I believe this is similar for the relationship between effectiveness and the substantive goals of EU law: Justin Lindeboom, Interpreting the EU Internal Market, in The Internal Market and the Future of European Integration 81, 82–84, 104–109 (Fabian Amtenbrink et al. eds., 2019).

81 Similarly, Kelsen considered the choice between “national law monism” and “international law monism” arbitrary from the viewpoint of legal science (Kelsen, supra note 25, at 388). The Westphalian ideal of sovereign equality only offers an ethical benchmark. For an ethics of European integration in favor of EU law primacy, see generally George Letsas, Harmonic Law, in Philosophical Foundations of EU Law (Julie Dickson & Pavlos Eleftheriadis eds., 2012).