On Courts of Last Resort and Lenders of Last Resort

At the heart of the Gauweiler case on the European Central Bank’s (hereafter: the Bank) bond buying programme are two delicate and fundamental dimensions of the European constitutional order.¹ The first dimension concerns the autonomy of the European legal order, captured by the question of who ultimately decides on the validity and interpretation of EU law. The second constitutional dimension relates to the nature of the EU’s economic and monetary union and concerns the question of what, if any, the role of the Bank is as a lender of last resort for sovereigns in the euro area.

Courts of Last Resort

Procedurally, the decision of the European Court of Justice of 16 June 2015 in the Gauweiler case is a model of restraint. Compared to Opinion 2/13 of 18 December 2014 on the accession of the EU to the European Convention on Human Rights,² the judgment is admirably low-key, despite the clear analogy between the two cases: both are about the preservation of the autonomy of the EU legal order. In the context of Opinion 2/13 the threat, whether real or perceived, mainly comes from the world outside, as represented by the European Convention on Human Rights and more specifically the European Court of Human Rights. In the context of the Gauweiler decision it comes from within, from a constitutional court of a member state: the German Bundesverfassungsgericht.

The Court of Justice and the Bundesverfassungsgericht agree that the EU is more than an ordinary international organisation. They also agree that the EU is not a state. However, the two courts draw diametrically opposed conclusions from the latter premise. In the view of the Court of Justice, the EU’s lack of statehood makes its legal order vulnerable. Therefore, the Court demands extra protection for and makes an absolute law of the EU’s autonomy, defending it almost instinctively, as a mother protecting her child – and indeed, the concept is the Court’s own offspring. For the Bundesverfassungsgericht, the EU’s lack of statehood is the reason to put the EU’s autonomy, which it recognises to a certain extent, into perspective, and to claim the right to subject secondary EU law to ultra vires and constitutional identity review.

In Opinion 2/13, the Court of Justice opposed the threat in full battle dress and with overblown constitutional rhetoric, elaborating on the autonomy of the EU legal order and its particularities. According to the Court, the accession of the EU to the European Convention on Human Rights as envisaged by the draft agreement is liable to affect adversely the specific characteristics of EU law and its autonomy, including, for example, the preliminary ruling procedure provided for in Article 267 TFEU,³ and the obligation of mutual trust between the member states.⁴

In Gauweiler, any reference to this is conspicuous by its absence. This is all the more remarkable because several of the intervening member states had urged the European Court to dismiss the Bundesverfassungsgericht’s request as inadmissible: ‘a reference for a preliminary ruling is not a procedural mechanism to make it easier for national courts or tribunals to carry out their own review of the validity of EU acts’.⁵ Fortunately, the Court did not walk into this trap. Wisely, it also resisted the temptation to restate its classic claims regarding the autonomy and the primacy of EU law, whether primary or secondary, over all national law of the member states, including their constitutional law. Instead, the Court relied almost entirely on the force of its substantive arguments that the decision of the European Central Bank on the Outright Monetary Transactions programme (OMT) falls with the Bank’s mandate and is not ultra vires. It only permitted itself to remark, almost in passing, that ‘it is settled case-law of the Court that a judgment in which it gives a preliminary ruling is binding on the national court, as regards the interpretation or the validity of the acts of the EU institutions in question, for the purposes of the decision to be given in the main proceedings’.⁶

Karlsruhe is well aware of the Luxembourg case law and still does not fully adhere to it, and a firm restatement of the autonomy and primacy of EU law would not have changed that. While the expounding of the classic doctrines of EU law in Opinion 2/13 was necessary to underpin the unconstitutionality of the draft accession agreement, in the case at hand such an exposé would not only have been to no avail, but possibly also would have added fuel to the fire. In this respect the Court has read the situation better than Advocate General Cruz Villalón did in his Opinion. Discussing the Bundesverfassungsgericht’s claim to have the right to exercise a constitutional identity review, the Advocate General remarked that:

(…) the Court of Justice has long worked with the category of ‘constitutional traditions common’ to the Member States when seeking guidelines on which to construct the system of values on which the Union is based. Specifically, the Court of Justice has given preference to those constitutional traditions when establishing a particular culture of rights, namely that of the Union. The Union has thus acquired the character, not just of a community governed by the rule of law, but also of a ‘community imbued with a constitutional culture’. That common constitutional culture can be seen as part of the common identity of the Union, with the important consequence, to my mind, that the constitutional identity of each Member State, which of course is specific to the extent necessary, cannot be regarded, to state matters cautiously, as light years away from that common constitutional culture. Rather, a clearly understood, open, attitude to EU law should in the medium and long term give rise, as a principle, to basic convergence between the constitutional identity of the Union and that of each of the member states.⁷

Here, a clear misunderstanding of the concept of German constitutional identity is revealed. The quote is fitted to the concept of constitutional identity as developed by the French Conseil constitutionnel.⁸ At least in the dominant interpretation, l’identité constitutionnelle de la France is made up of French constitutional rules and principles which are not also part of the EU’s legal order, in other words of constitutional rules and principles which are unique to France, such as laïcité, the French version of the separation of church and state, and the prohibition on giving specific rights to ethnic or cultural minorities.⁹ And Cruz Villalón is right: it is quite possible that the years to come will take the edge off constitutional differences between the member states and the Union in this respect.

However, Germany’s constitutional identity refers to principles which are not only foundational for the national legal orders of the other member states, but also for the EU’s legal order itself: democracy, rule of law, fundamental rights protection, solidarity (Articles 2 and 3 TEU). In this perspective, the need to protect German constitutional identity vis-à-vis the EU is not diminished by the identity, in the sense of sameness, of their foundational principles. German constitutional identity requires that until the German people as the original holder of German sovereignty in free self-determination decides otherwise, Germany remains a democratic, rule-of-law-based, fundamental rights-protecting, social and federal state in its own right. This is, mutatis mutandis, a view taken by several other national constitutional courts too. Recently, in the Pham judgment of the United Kingdom Supreme Court of 25 March 2015 dealing with the possible application of European law to an order depriving Mr Pham of British citizenship (rendering him stateless), Lord Mance expressed their common basic idea very eloquently:

For a domestic court, the starting point is, in any event, to identify the ultimate legislative authority in its jurisdiction according to the relevant rule of recognition. The search is simple in a country like the United Kingdom with an explicitly dualist approach to obligations undertaken at a supranational level. European law is certainly special and represents a remarkable development in the world’s legal history. But, unless and until the rule of recognition by which we shape our decisions is altered, we must view the United Kingdom as independent, Parliament as sovereign and European law as part of domestic law because Parliament has so willed. The question how far Parliament has so willed is thus determined by construing the 1972 Act.¹⁰

On this basis, Lord Mance also very clearly claimed the right for British courts to exercise ultra vires and constitutional identity review:

But, unless the Court of Justice has had conferred upon it under domestic law unlimited as well as unappealable power to determine and expand the scope of European law, irrespective of what the Member States clearly agreed, a domestic court must ultimately decide for itself what is consistent with its own domestic constitutional arrangements, including in the case of the 1972 Act what jurisdictional limits exist under the European Treaties and upon the competence conferred on European institutions including the Court of Justice.¹¹

To return to German constitutional identity and to the OMT decision of the European Central Bank: until the ultimate rule of recognition in Germany is altered, the ‘haushaltpolitische Gesamtverantwortung’ of the Bundestag, the overall budgetary responsibility of the German parliament, is integration-proof and may not be eroded by Germany’s participation in the monetary union.¹² And until that decision is taken, the Bundesverfassungsgericht is required to exercise ultra vires and constitutional identity review, in order to prevent the Union institutions, in their natural quest to maximise their powers, from eroding the basic necessities for a viable German political community: the Bundestag needs to retain substantial budgetary (and legislative) competences of its own in order for Germany to be able to ‘democratically shape itself.’¹³

The decision to change the ultimate rule of recognition is, to borrow Cruz Villalón’s metaphor once again, most probably still light years away. Until then, the Union will have to cope with different courts, which on the basis of different rules of recognition claim to be the court of last resort. In such a situation, where parties cannot settle their differences of opinion by reference to an ultimate worldly authority recognised by both, and the use of force is excluded, there is only one remedy: the recourse to reason and arguments to convince the other. And that is precisely what the Court has done in Gauweiler.

Lenders of Last Resort

The financial and euro area crises have painfully illustrated the consequences of the lack of a credible fiscal backstop for sovereigns in the euro area.¹⁴ Many countries have a traditional lender of last resort in the form of the central bank, ready to take responsibility for preventing sovereign default, willing to implicitly guarantee public debt by making it known that, if necessary, unlimited financing capacity will be mobilised. The striking thing about this message is that, if it is credible, the chances of the capacity actually being mobilised are very small.

The constitution of economic and monetary union was, however, not built to follow this logic. Not only did the legal framework of the EU, based on the fundamental principle of attribution of powers, not provide for a lender of last resort function, neither with the European Central Bank nor with the euro area member states collectively, but two central provisions were also included in the economic and monetary union with the aim of preventing this function, namely Article 123 TFEU (often labelled the prohibition of monetary financing) and Article 125 TFEU (often referred to as the no bail-out clause).

The crisis has forced the euro area to come to terms with its own reality. The idea of perpetual capital market access for member states at tolerable interest rates has proven to be an illusion. The same is true for the idea that member states of the euro area will not default. Responding to this reality has been, and still is, a tremendous political, economic and legal struggle. From an economic and political perspective, two key moments are so far among the most decisive. The first is the weekend of 9/10 May 2010, when Europe’s political leaders decided to set up temporary emergency funds capable of providing emergency assistance to member states (later followed by the permanent European Stability Mechanism (ESM)),¹⁵ and the European Central Bank in parallel decided to start buying government bonds through its first bond buying programme (called the SMP-programme or Securities Markets Programme).¹⁶ The second moment was on 26 July 2012, when President of the European Central Bank Mario Draghi made the following announcement: ‘Within our mandate, the ECB is ready to do whatever it takes to preserve the euro’.¹⁷

For a legal reading of these events, two key moments are the following. First, the Pringle judgment of the Court of Justice on 27 November 2012 on the European Stability Mechanism (ESM) Treaty.¹⁸ And second, the Court decision of 16 June 2015 in the Gauweiler case,¹⁹ on the main policy expression of the Bank’s ‘whatever it takes’ message: the second bond buying programme announced in a press release of September 2012 (called the OMT programme).²⁰

The way that the European legal order deals with the issue of lender of last resort is central to both Pringle and Gauweiler, albeit in different ways. The legal question at the heart of Pringle is whether the EU treaties permit the creation of the ESM emergency fund outside the framework of the Treaties. The Court of Justice decided that this is permitted, since EU law does not, in Article 122 TFEU, provide a legal basis for the establishment of a stability mechanism of the kind of the ESM, and because the Union is not obliged to use Article 352 TFEU for that purpose.²¹ Providing support through the ESM is, however, only compatible with Article 125 TFEU if it is indispensable to safeguard the financial stability of the euro area as a whole and of its member states, and if it is subject to strict conditionality.²²

Thus, as long as these two conditions are met, the EU legal framework allows the euro area member states to create a vehicle capable of providing unlimited assistance in order to prevent sovereign default. The political reality, however, is such that the member states are not willing to provide unlimited assistance, and the firepower of the ESM is in fact limited to a maximum lending volume of €500 billion. This, it has been argued, would not be sufficient to save Italy and Spain, undermining the credibility of the ESM as a lender of last resort.

And so, in the summer of 2012, at a time when interest rates on bonds of Italy and Spain were arguably unsustainably high, due to ‘ungrounded fear’ of and speculation on a breakup of the euro area, the Bank’s President, Draghi, came with his famous ‘whatever it takes’ message that many have interpreted as the announcement of a willingness to act as a lender of last resort. Economists argue that the Bank’s sovereign bond buying programmes testify that it has now de facto become a lender of last resort for sovereigns.²³ In fact, the similarities between the OMT-programme – a commitment to buy sovereign bonds without an ex ante quantitative limit – and the lender of last resort function are evident, especially in the context of Draghi’s promise to do ‘whatever it takes’ within the Bank’s mandate to save the euro. But what does the law say? For this we return to the Gauweiler case, at the heart of which is, again, the relationship between EU law and the issue of lender of last resort.

In Gauweiler, the central legal question is whether EU law allows the Bank’s OMT bond-buying programme. In its preliminary referral, the Bundesverfassungsgericht has shown itself to be extremely critical of the OMT programme: the OMT programme is an economic policy act equivalent to an assistance measure,²⁴ and it goes beyond the Bank’s mandate. It is aimed at neutralising spreads on government bonds,²⁵ and it violates the monetary financing prohibition, among other things because of the absence of an ex ante quantitative limit. In other words, according to the German court, through the OMT programme the Bank is acting as lender of last resort, which is prohibited by EU law.

The Court of Justice takes a fundamentally different view from the German constitutional court. OMT is a proportional monetary policy measure, and one that falls within the Bank’s mandate in terms of both instrument and objective,²⁶ since ‘measures that are intended to preserve that transmission mechanism may be regarded as pertaining to the primary objective’ of price stability.²⁷ The Court grants the Bank a broad margin of discretion in conducting monetary policy, because it has to make choices of a technical nature and to undertake forecasts and complex assessments.²⁸ This places a considerable amount of trust in the Bank.²⁹ The Court is aware of the fact that the instrument of OMT, namely sovereign bond buying on the secondary market, may also be employed by the ESM, but considers it to be a decisive factor that the objective of the two instruments is different: the OMT programme may only be implemented in so far as it is necessary for the maintenance of price stability (a monetary policy objective), while the ESM’s intervention is intended to safeguard the stability of the euro area (an economic policy objective).³⁰ In reality, of course, the two objectives seem to be insolubly linked: if there is serious fear on the financial markets for a break-up of the euro area, the spread between interest rates of government bonds in the euro area and the rates of the bonds of the weak member states will inevitably rise, obstructing in turn the Bank’s monetary policy transmission channels.

On the important absence of an ex ante quantitative limit, which is arguably also the main sting in the Bundesverfassungsgericht’s assessment that OMT potentially involves German constitutional identity, the European Court takes a very pragmatic approach and makes the following relevant remarks. First, the potential scale of the programme is de facto limited in a number of ways. Second, the Bank could legitimately decide not to adopt a quantitative limit to the OMT programme prior to its implementation, as such a limit would likely reduce the programme’s effectiveness. Third, the de facto restriction of the volume of government bonds eligible to be purchased in the framework of the programme leads to only a limited impact on the impetus for member states to follow a sound budgetary policy. This adds to the Court’s conclusion that OMT does not lead to a circumvention of the prohibition of monetary financing of Article 123 TFEU. Finally, according to the Court, the financial risks of the OMT programme are simply inherent in monetary policy.³¹ Legally, one of the intriguing remaining questions is whether an OMT programme with the same objective to unblock the Bank’s monetary policy transmission channels would still pass the proportionality test if it were de facto unlimited.

How, then, to translate the Gauweiler judgment in terms of the lender of last resort function? Article 123 TFEU does not prevent sovereign bond buying on the secondary market, not even through a programme that sets no ex ante quantitative limits. Importantly, the European Court of Justice is not saying that the Bank would be allowed to act as lender of last resort for sovereigns. Instead it can be interpreted as saying that, from a European law perspective, the Bank, with its OMT programme, is not acting as one. The Outright Monetary Transactions programme, even though it consists of sovereign bond buying, is not an act of financial assistance but an instrument of monetary policy. Even if it has indirect effects on the stability of the euro area, this does not make it an act of economic policy in terms of EU law.³² The marginal review of the proportionality of monetary policy measures gives the Bank a broad discretion, and these measures may have (according to the Court: will actually always have)³³ an indirect effect on the financing conditions of member states without therefore necessarily being in violation of the prohibition of monetary financing. In short, monetary policy may have the indirect effect of stabilising the government bond market.³⁴

All this leads to an interesting conclusion, especially for lawyers. No quantitative limits exist under EU law to a(n external) lender of last resort function carried out by the member states of the euro area collectively, but there is not enough political will to exercise it convincingly through the ESM. By contrast, from an EU law perspective, the Bank is not acting as a lender of last resort in its OMT programme, and EU law does not provide it with a mandate to act as one either.³⁵ Again, however, this legal dimension does not tell us the whole story, as the confidence of bond holders in the existence of a perceived lender of last resort is not only determined by a legal reading of the OMT events.

Thus, it seems that so far the Bank manages to successfully speak two languages to its different audiences: lawyers and bondholders. The Court is satisfied with the current design of the OMT programme, and with the focus of the Bank on the practical limits of the programme, in view of the prohibition of monetary financing. Bondholders, by contrast, are satisfied with the ‘whatever it takes’ message, combined with the absence of an ex ante quantitative limit to the size of the bond purchases. In fact, the room for manoeuvre which the Court of Justice grants the Bank in Gauweiler will only strengthen the perception of the bondholders that the Bank indeed can and will act as a lender of last resort. Arguably the Bank’s new, third sovereign bond-buying programme (called the Public Sector Purchase Programme) points in the same direction.³⁶ This latest programme, announced in January 2015, is generally understood as the Bank embarking on quantitative easing, as it will buy government bonds on the secondary market, worth up to €60 billion, each month until at least September 2016. Interestingly, with the Public Sector Purchase Programme the Bank illustrates that massive government bond buying has now become a monetary policy instrument not confined to member states in trouble.

The OMT programme, with its indirect effects on the stability of the government bond market, in combination with the ESM emergency fund, testifies to the gradual and combined development of a lender of last resort function at the supranational level. This includes the explicit and limited lender of last resort function of the ESM, and the implicit one of the Bank’s ‘whatever it takes’ message. Responsibilities and burdens are therefore shared. A central part of this gradual development is the effort to address the so-called moral hazard problem: how to neutralise the incentive to governments to issue too much debt.³⁷ In fact, ESM assistance is combined with strict conditionality, activation by the Bank of the OMT programme is conditional upon ESM assistance, and the availability of quantitative easing through the Public Sector Purchase Programme for member states that receive financial assistance is dependent on a positive outcome of a review by the ‘institutions’ (better known as the Troika).³⁸ Related to this, the most recent events with regard to Greece illustrate further political limits to the lender of last resort function. Both the (other) member states of the euro area and the European Central Bank have shown great reluctance to provide direct or indirect liquidity, in the absence of credible political willingness or factual incapability on the side of the recipient state (Greece) to safeguard the sustainability of public finance.

In part, these developments relate to the overall reinforcement of the economic and monetary union.³⁹ In part they also simply put into perspective the legal reality of the Gauweiler case, and of economic and monetary union at large.

If economic reality serves to put a legal reading of OMT in perspective, OMT does the same with a certain economic-political reading of the Treaties. In the same way as the Pringle case did, Gauweiler painfully highlights the mismatch between the framing of important Treaty provisions on the economic and monetary union on the one hand, and the details of the legal interpretation of these provisions given by the member states and institutions, and taken over by the Court of Justice, on the other. Labelling Article 125 TFEU the ‘no-bailout provision’ as opposed to the ‘provision prohibiting some but not all forms of financial assistance’ is not merely a question of linguistic convenience. It also expresses a specific understanding of the provision and serves as a signal to the markets. The same is true for speaking of a ‘prohibition of monetary financing’ as opposed to a ‘prohibition to buy government bonds on the secondary market only if they have the same effect of a purchase on the primary market or lessen the impetus of the member states concerned to follow a sound budgetary policy’. Both Pringle and Gauweiler show the tension between conventional wisdom on economic and monetary union and the capacity of EU law to accommodate new developments and events, in particular in times of crisis.⁴⁰ This may be a tough lesson to learn for – especially German – economists who support only a specific understanding of economic and monetary union. Incidentally, this conventional understanding of economic and monetary union referred to is also, to a certain extent, dictated by German constitutional law and even German constitutional identity.⁴¹

It is now up to the Bundesverfassungsgericht to taste the Gauweiler judgment.

Should it come to an ultra vires conclusion, open conflict would follow. Should it also conclude that it affects German constitutional identity, things would be even more complicated. Both decisions would put the indirect ‘lender of last resort’ effects of OMT in a new perspective. The Bundesbank, or any other German authority, would not be allowed to take part in its implementation.⁴² Could OMT (and the EMU) survive such a legal and political blow? From a German perspective, in case of an ultra vires conclusion, OMT could be redeemed by a Treaty amendment which, for Germany, is approved by the German parliament, if need be with a two-thirds majority (Article 23(2) Grundgesetz); in case of a violation of constitutional identity, it is required to obtain in some way the approval of the German people. Would OMT be worth that trouble? If the stabilising function of Bank measures on the sovereign bond market could probably survive in some other form as well, the consequences for the cooperative relationship between the Court of Justice and the Bundesverfassungsgericht would be much more dramatic after Luxembourg’s intra vires call.

The German Court might of course also go along with the Gauweiler ruling. At least in terms of time sequence, in the chain of events that constitutes the dance of justice between courts in Europe,⁴³ the national constitutional court is the court of last resort. This makes it more difficult to disagree with the European Court of Justice, but the agreement is ultimately its own choice.

TWB/JHR