25 Mai 2020

The Bundesbank is under a legal obligation to ignore the PSPP Judgment of the Bundes­verfassungs­gericht

If there is a situation undermining the rule of law, then it is exactly this: The Bundesbank is under a legal obligation to ignore the PSPP Judgment of the Bundesverfassungsgericht (under EU law), and the Bundesbank is under a legal obligation to follow the PSPP Judgment of the Bundesverfassungsgericht (under German constitutional law). How has it come to this?

When confronted with the question of the supremacy of EU law over national constitutions, the ECJ on the one hand and national constitutional courts, especially the Bundesverfassungsgericht, on the other, have always represented irreconcilable positions. Nevertheless, they managed to prevent a clash that would have damaged the entire integration process. This was not just mere judicial courtesy: the existing détente reflected the reality that none of the actors could claim to be ultimately right simply because deciding between right and wrong is dependent on the point of departure one is choosing. The positions are well known and have not changed since the beginning.

1. The ECJ Perspective

For the ECJ, supremacy has been a direct and inevitable consequence of the special nature of EU law at the outset. Already in Costa v ENEL the ECJ made it clear that EU law “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 and without the legal basis of the community itself being called into question” (emphasis added). It is by no surprise that the exact same wording is used in Internationale Handelsgesellschaft (para. 3) to justify the supremacy of EU law over national constitutions. Later cases involving a conflict of EU law with national norms of constitutional rank no longer used the specific term “nature” (Factortame II (para. 22): effectiveness of EU law, Melloni (para. 63): efficacy of the framework decision), but the underlying idea has remained the same: without full and unconditional supremacy EU law would no longer be what it is supposed to be. 

Naturally, in order to immunise the EU legal system from national constitutional challenges, its special nature must inevitably encompass its autonomy from the domestic legal systems. Only if EU law and EU law itself can determine the rules pertaining to its relationship with the domestic legal system, can national constitutional reservations be deemed irrelevant in case of a conflict. Such autonomy has been stipulated right from the beginning, albeit without elaborating its meaning. Its first mention can be tied already to the era of the European Coal and Steel Community, where advocate General Legrange’s 1956 Opinion in Mirrosevich v High Authority (ECLI:EU:C:1956:9, p. 399), described the nature of the law of ESCS as possessing an autonomous character. While the original French version used the term ‘autonome’, it was translated to English as ‘independent’, taking the edge out of the term. In Van Gend en Loos the ECJ refrained from using the term autonomy and referred rather to “a new legal order of international law”, but Costa v ENEL and Internationale Handelsgesellschaft were based on the premise that the founding treaties of the EEC emanate from an ‘independent source’ – “une source autonome”. From time to time ever since, the ECJ confirms (here at para. 166) that autonomy and supremacy are specific characteristic arising from the very nature of EU law.

2. The Bundesverfassungsgericht Perspective

The Bundesverfassungsgericht has remained more subtle in its argument and has not so far invoked the nature of the Grundgesetz as ‘constitution’ expressly. Nevertheless, the underlying premise does follow from the supposed characteristic of the constitution being the sole source of all public authority for those under the jurisdiction of the Federal Republic of Germany. This characteristic is necessary to postulate the theory of constitutional authorisation (Theorie der verfassungsrechtlichen Ermächtigung), a theory on which all constitutional reservations towards EU law are based. Only if we accept that sovereign powers can be exercised in Germany solely on the basis of the Grundgesetz, can we claim – as the Bundesverfassungsgericht does – that EU law is applied because the Grundgesetz commands so. In other words, the gatekeeper function of the Grundgesetz follows directly from the nature of the same Grundgesetz being the foundation of all public authority in Germany.

This character of a constitution might seem self-evident in a state-centred constitutional theory, nevertheless it runs counter the realities of a supranational organisation. What is more, it does not even follow from the wording of Article 24 of the Grundgesetz, the integration clause in force at the time the fundamental rights reservation was first framed in Solange I. On the contrary, as recognised by the Bundesverfassungsgericht in Solange I (at para. 43), Article 24 (1) of the Grundgesetz expressly provides for the possibility of transfer of ‘sovereign powers’ (“Hoheitsrechte übertragen”) to international organisations. In fact, the Bundesverfassungsgericht had to go at length to argue that the language of Article 24 (1) GG must be interpreted in the context of the entire constitution, already then referring to the identity of the Grundgesetz. This interpretation allowed for the conclusion that “Article 24 GG does not actually authorise the transfer of sovereign powers, rather it opens the national legal order (with the stated limitations).” This denial of the possibility of actual transfer of sovereign powers has remained the basis of all future constitutional reservations, even though the language of the new integration clause, Article 23 introduced in 1992 equally authorises the transfer of sovereign powers. The Lisbon Judgment (at para. 226 ff.) attempts to distinguish between sovereign powers (“Hoheitsrechte”) and sovereignty, but this distinction fails to explain why the autonomous EU legal system shall remain “derived” (abgeleitet) (at para. 231) in the sense that it cannot take precedence over the national constitution in those areas where sovereign powers have already been transferred. 

The comprehensive and exclusive nature of the constitution and its resulting gatekeeper function vis-à-vis EU law allows for transforming questions of interpretation of EU law into constitutional questions, a keen aspiration of the Bundesverfassungsgericht ever since Maastricht. The ultra vires reservation triggered in the PSPP judgment is doing exactly this. Even if the boundaries of the competences of EU institutions follow from the founding treaties, the Bundesverfassungsgericht claims the power to ultimately adjudicate about these because, in the eyes of Karlsruhe, it is ultimately a German constitutional issue. The argument leading to this conclusion through the right to vote as well as the principles of democracy and popular sovereignty is more than complicated (to put it mildly), especially in relation to an independent institution, like the ECB. But even if (supposedly) we are ready to accept the twists and turns of the reasoning of the PSPP judgment, they only hold true if we also accept that Germany never gave away sovereign rights, only competences ultimately limited by the Grundgesetz.

The above is not to say that the theory of constitutional authorisation is per se false, untrue or illogical. And we cannot say that about the autonomist view either. Leaving aside the actual German PSPP judgment and the respective ECJ judgment, both positions could be doctrinally coherent if we accepted their respective premises. The choice between the two perspectives, at the end of the day, rests on a Werturteil in a Weberian sense, or on a political choice just like Kelsen described the choice between primacy of international law and primacy of domestic law.

3. The Lack of Mutually Accepted Rules on Norm Collision

That no actual conflict has erupted from these irreconcilable positions can be thanked to the clever judicial politics of Karlsruhe and Luxemburg which tried to accommodate the basic needs of the competing actor without compromising their own theoretical foundations. The best and probably most beneficial example of this wisdom is the Solange II judgment of the Bundesverfassungsgericht, which also became the model of the Bosphorus presumption governing the relationship of EU law and the ECHR.

The conflict can doctrinally be conceptualised either from an ECJ perspective or from a Bundesverfassungsgericht perspective. From the ECJ perspective, there is a clear doctrinal answer: as the PSPP judgment of the Bundesverfassungsgericht is itself just a national legal act, in case of conflict (which is obviously given here) it is inapplicable, therefore the Bundesbank clearly has to ignore the Bundesverfassungsgericht judgment (and even an infringement procedure is possible against Germany). Also from the Bundesverfassungsgericht perspective, there is a clear doctrinal answer: the PSPP judgment has clarified the limits of the supremacy of EU law (again), so the respective ECJ judgment is inapplicable, therefore the Bundesbank clearly has to ignore the ECJ judgment. The respective collision rules of EU law and German law have always contradicted each other, there have never been mutually accepted (doctrinal or codified) rules on norm collision.

All eyes are on the Bundesbank now. In its first reaction, the Bundesbank expressed hopes that the problem will just go away and the ECB will do what is required by the Bundesverfassungsgericht. The ECB, however, seems totally unimpressed and considers the German judgment as non-binding. Time is ticking for the Bundesbank: the most likely scenario at the moment when we are writing this blogpost is that the three months deadline (which was set by the Bundesverfassungsgericht) will pass, and the Bundesbank will have to make a choice. Whatever the Bundesbank will do, it will be unlawful: exiting PSPP will breach EU law, staying in it will breach German law. But even if for some miraculous reason the actual collision situation could be avoided (a sudden major reform finishing PSPP), the old paradigm about handling norm collisions is gone. 

The question, whether the Bundesverfassungsgericht judgment was doctrinally well-argued is equally irrelevant for this problem (for convincing sceptical voices, see here, here and here), just like it is irrelevant whether the former ECJ judgment was doctrinally well-argued. The issue is not how convincing the reasoning in any of the judgments is. This is now about the rule of law in a situation where legal obligations conflict each other at the highest level without mutually agreed collision rules. In this context, the Bundesverfassungsgericht’s step looks more like starting a chicken game.

The PSPP judgment is not a new turn primarily because it would invent a new doctrine or because it would deviate from the existing case law (even though some doctrinal steps in the reasoning are quite surprising: we would like to avoid the stark terms that the Bundesverfassungsgericht used about the ECJ judgment). Nor it is new in the sense the Bundesverfassungsgericht never forbade to fulfil an obligation under EU law: this boundary was already crossed in 2015 in relation to the execution of a European arrest warrant. It is a turn mainly because it is flagrantly giving up on a paradigm that ensured peace between two of the most important courts in Europe. The paradigm that maintaining one’s theoretical position can be combined with a practical way of living together. The paradigm that the rule of law is diminished if state agencies are exposed to contradicting legal obligations both seeking to be supreme over the competing one, and the agencies are forced to make a choice that will inevitably be illegal, one way or another. Duly noted, this paradigm has never been of doctrinal nature. It was clever judicial politics in order to preserve important constitutional values and European integration at the same time. It is unclear, what strategic goal the Bundesverfassungsgericht wanted to achieve with the PSPP judgment. It is, however, more than likely that this new sort of judicial politics will equally hurt European integration and the rule of law. 

SUGGESTED CITATION  Jakab, András; Sonnevend, Pál: The Bundesbank is under a legal obligation to ignore the PSPP Judgment of the Bundes­verfassungs­gericht, VerfBlog, 2020/5/25, https://verfassungsblog.de/the-bundesbank-is-under-a-legal-obligation-to-ignore-the-pspp-judgment-of-the-bundesverfassungsgericht/, DOI: 10.17176/20200526-013433-0.


  1. Jörn Sack Mi 27 Mai 2020 at 13:41 - Reply

    I am surprised that a law professor ignores largely the relevant Art. 23 GG (Art. 24 is relevant for other international organisations than the EU) and gives no consideration to Art. 79 III GG which the BVerfG invoked and to which Art. 23 refers.

    • Tataru, Mihai Coriolan Do 28 Mai 2020 at 09:55 - Reply

      He referres to the old Art. 24 GG, which was the integration clause that is now Art. 23

      • Nathan de Arriba-Sellier Sa 30 Mai 2020 at 08:09 - Reply

        Interesting blog, which contributes to a much needed discussion on what the Bundesbank should do after the PSPP decision. Although I share the opinion of the authors, it remains, however, still a little bit unclear to me how they reach the conclusion that the Bundesbank is under a legal obligation to ignore the Bundesverfassungsgericht on that matter. This would have probably needed longer developments. While the conflict of legal orders makes the general question very difficult to solve, I believe that the answer, in the specific case of monetary policy, is already to be found in the law (https://eulawlive.com/op-ed-between-karlsruhe-and-luxembourg-lies-frankfurt-the-bundesbank-and-the-bundesverfassungsgerichts-pspp-decision-by-nathan-de-arriba-sellier/).

        • Daniel Reichert-Facilides Mi 10 Jun 2020 at 21:53 - Reply

          I only half agree: while the Bundesbank is certainly subject to the ECJ jurisdiction under Art. 271 d) TFEU, the acting civil servants are primarily bound by the German constitution (as interpreted by the BVerfG) under § 31(3) BBankG and §§ 60(1) and 64(1) BBG. By ignoring the PSPP decision, they would (at least in theory) risk disciplinary action.

          But that does not mean that the Bundesbank would get away with not participating in further PSPP transactions after 5 August 2020, even if the ECB did not pass another resolution on proportionality: in this scenario, the ECB could (simply) step in for Bundesbank under Artt. 3 and 9.2 of the ECB Statute, and Bundesbank would be required to account for its share in the transactions under § 26(1) BBankG.

  2. Balazs Koranyi Do 28 Mai 2020 at 10:50 - Reply

    The judgment requires the Bundestag and the Federal Government to ensure compliance with proportionality. So ultimately, they hold the key to diffusing this tension. If they accept the Bundesbank’s assessment, they’ve complied with the BVerfG ruling and the Bundesbank is off the hook. Of course, their decision can be challenged in court again, but, in this case, it’s the Bundestag and the Government that may be breaking the court ruling and not the Bundesbank, which is probably easier to swallow. The only sticking point is that the BVerfG required the ECB to take a new decision. This is problematic because the Governing Council will not take an order from a national court. But a legal trick can solve this. An option may be that the Bundesbank prepares an assessment and Jens Weidman – so not the Court – makes a request to the GC to review the document. The GC then concludes, without any reference to the German Consitutional Court, that the Buba assessment is fair.

  3. Inga Feldmann Mo 30 Nov 2020 at 17:02 - Reply

    Just a short note: the first link for PSPP judgement leads to the 30 July 2018 ruling on the ECB banking supervision

Leave A Comment


1. We welcome your comments but you do so as our guest. Please note that we will exercise our property rights to make sure that Verfassungsblog remains a safe and attractive place for everyone. Your comment will not appear immediately but will be moderated by us. Just as with posts, we make a choice. That means not all submitted comments will be published.

2. We expect comments to be matter-of-fact, on-topic and free of sarcasm, innuendo and ad personam arguments.

3. Racist, sexist and otherwise discriminatory comments will not be published.

4. Comments under pseudonym are allowed but a valid email address is obligatory. The use of more than one pseudonym is not allowed.

Explore posts related to this:
BVerfG, Bundesbank, Constitutional Pluralism, ECJ, PSPP, Ultra Vires

Other posts about this region:
Deutschland, Europa

Explore posts related to this:
BVerfG, Bundesbank, Constitutional Pluralism, ECJ, PSPP, Ultra Vires

Other posts about this region:
Deutschland, Europa
6 comments Join the discussion