1 - The Constitutional Boundaries of European Fiscal Federalism

Summary

Chapter 1 evaluates competing EU and member state claims over the constitutional boundaries of ‘constitutional identity’ and Kompetenz-Kompetenz in the European legal order, and identifies two constitutional boundaries of the EU legal order which will impinge on the selection of fiscal federalism models for the EU. The first is fiscal sovereignty. Member state fiscal sovereignty is a permanent constitutional constraint on the application of fiscal federalism theory in the EU. The Union can have no powers other than what the member states have given it, and nemo plus iuris, what the member states have given it is limited by their own ‘constitutional identities’ – of which fiscal sovereignty lies at the core. The second is comprised of the fundamental guiding principles of price stability and ‘fiscal discipline’ (sound public finances, and sustainable balance of payments) binding on the mandate for EMU itself under Article 119(3) TFEU.

This chapter introduces and establishes the constitutional boundaries of European fiscal federalism which are the object of this book:

• [1.3.1] The first constitutional boundary of European fiscal federalism studied in this book is Member State fiscal sovereignty. Fiscal sovereignty, broadly defined, refers to the exclusive competences of national legislative organs for economic and fiscal policy as they are charged with those competences by the pouvoir constituent. Economic and fiscal policy competences comprise the ‘core of parliamentary rights in democracy’ and a material limit of ‘constitutional identity’ in Europe’s twenty-eight constitutional democracies.¹ According to European constitutional identity and Kompetenz-Kompetenz jurisprudence, a deprivation of fiscal sovereignty would require Member States to repudiate encroaching EU law (refusing to apply the offending EU instrument) or withdraw from the Union altogether.² This chapter tests the veracity of this constraint as a constitutional boundary of EU fiscal federalism and extracts three tests for evaluating whether a proposed legal model infringes Member State fiscal sovereignty. These are: no unlawful restrictions on fiscal sovereignty;³ no unlawful conferral or delegations of fiscal sovereignty;⁴ and no structural impairments of fiscal sovereignty through financial dispositions of structural significance to budgetary autonomy.⁵

• [1.3.2] The second constitutional boundary is comprised of the fundamental guiding principles of price stability, sound public finances and a sustainable balance of payments enshrined in the mandate for EMU under Article 119(3) TFEU. It is these principles which form the basis of Member State (in particular, German) acts of accession, and it is these principles to which the entire legal architecture of EMU under Articles 119–127 TFEU is attuned.

However, before the constitutional boundaries which bear upon the field of fiscal federalism can be established, it must first be established that there are, indeed, constitutional boundaries which constrain the expansion of the EU legal order as a whole. This is so because, as a matter of pure EU law, the boundaries of the EU legal order are limitless in their potential. The scope of EU law is set out by the Treaties, and there are no substantive constraints on the amendment of those Treaties.⁶ From the internal perspective of the EU legal order, any model of federalism is compatible with EU law de lege ferenda upon the flourish of twenty-seven (formerly twenty-eight) pens.

Then, once a competence has been conferred on the Union, the ECJ has, since Costa v. ENEL and Internationale Handelsgesellschaft, declared that EU law has absolute primacy over all constitutional laws and structures of the Member States.⁷ National law must be interpreted in conformity with EU law,⁸ and, where they are in conflict, EU law must prevail.⁹ Secondary instruments such as regulations,¹⁰ directives¹¹ or decisions¹² will prevail over national constitutional or statute law, even if the national law is later in time.¹³ The CJEU is the sole arbiter of the legality of all EU measures, and it reserves for itself final authority to deliver binding rulings on the compatibility of EU law with fundamental rights and principles.¹⁴ As Claes so puts it, EU law requires national courts ‘to refrain from enforcing the constitutional provisions that they have a sword duty to uphold and protect, in favour of any act of Community law, whatever its rank or content’.¹⁵

Constituent within this supremacy claim is what is referred to in this study as the claim of ‘absolute’ supremacy: Not only does the CJEU determine the status and effect of EU law within its established competences (ordinary supremacy), but the CJEU is the final arbiter of the boundaries between Member State and EU competence (absolute supremacy).¹⁶

The question for the architects of EU fiscal federalism is whether this provides a true account of European constitutional law, or whether national legal orders are indeed capable of imposing constitutional constraints upon the selection of fiscal federalism models for the EU. In pursuit of that question, this chapter evaluates the competing claims of EU and Member State constitutionalism against three approaches to the validity of law in European legal theory:¹⁷ Pure (Kelsenian) constitutional theory;¹⁸ normative constitutional pluralism;¹⁹ and (Hartian) legal positivism.²⁰

Section 1.1 begins by familiarizing the reader with European constitutional theory and the competing claims of Member State and EU constitutionalism. What is normatively at stake in this dispute is the locus of sovereignty and therefore the question of Kompetenz-Kompetenz – that is, who is the ultimate arbiter of which competences have and have not been conferred on the Union. The analysis seeks to inform the architects of fiscal federalism on where they should look for an authoritative description of what is and is not safe constitutional ground when selecting from models known to fiscal federalism theory. Section 1.1 sets out the background explaining why this book finds it necessary to look to both EU and Member State law in doing so.

Section 1.2 shows that national constitutional orders profess to impose two types of limitation on EU law: First, Member State courts profess that they have the jurisdiction to assert, through treaty ratification and ultra vires review, what powers they have and have not conferred on the Union – the so-called Kompetenz-Kompetenz. Second, Member State courts assert that their own state ‘constitutional identities’ determine the absolute limits of conferral and application of Union law – the so-called ‘constitutional identity’ review jurisdiction.²¹ Section 1.2 evaluates the veracity of these claims as a matter of pure constitutional law, as normative legal principle, and as a positivist statement of law. It finds that, by all three approaches, these jurisdictions provide a valid description of the limits of the EU legal order for the purposes of this study. Of the twenty-eight Member States surveyed in this book, all assert Kompetenz-Kompetenz and twenty-four have developed a body of jurisprudence surrounding ‘constitutional identity’ – a set of constitutive principles so integral to the constitutive nature of the state that they are beyond the reach of the national (and European) legislator.

Section 1.3 conducts the main task of this chapter: To identify those constitutional boundaries which bear upon the field of fiscal federalism. It sets out those principles and tests which constitutional courts (and this book) will apply to novel legal apparatus in the field of fiscal federalism.

1.1 An Introduction to European Constitutionalism

1.1.1 European Monist Federalism and the Principle of Supremacy

The European Union is founded on the principle of democracy.²² An essential precept common to the legal heritage of the Member States is that the bearer of sovereignty is the people.²³ Under European ‘social contract’ theories of constitutionalism, the locus of sovereignty is indivisible.²⁴ At the base of every legal order is a historically first constitution – a revolutionary act – which is enacted by the pouvoir constituent originaire in a manner different from that prescribed by any prior constitution. This is Kelsen’s ‘basic norm’ (or Grundnorm) which forms the basis for the legal system.²⁵ Under European constitutional theory a ‘Union of States’ must, therefore, either be a ‘confederation’ (under which participants retain their character as sovereign states) or a sovereign ‘federal state’ (under which powers are devolved by the central government).²⁶ In a conflict of norms, only one institution can have the ultimate claim to empowerment by the pouvoir constituent.²⁷ Schütze explains:

Within this European tradition, ‘federalism’ came thus to refer to the constitutional devolution of power within a sovereign nation. A federation was a Federal State.²⁸

Coloured by this tradition, European constitutionalism from the 1960s treated the residual existence of Member State sovereignty as incompatible with EU federalism. The object of European law, namely, ‘to substitute a common and uniform European law for the divergences and conflicts of national bodies of legislation’,²⁹ required early European jurists to free it from the obvious criticism that there could be no such thing as an autonomous legal order superior to the Member States.³⁰ As Schütze so puts it, ‘[i]t became the task of European scholarship to make the “Federal State” look like its unitary sisters […] through feats of legal “reasoning”’.³¹

In Van Gend en Loos, and Costa v. ENEL, the ECJ famously stated that through ‘the establishment of institutions endowed with sovereign rights’ the Community constituted an ‘autonomous legal order’ stemming from ‘an independent source of law’.³² By asserting that ratification of the Treaties was a constituent act, a historically-first basic norm for a ‘constitutional charter based on the rule of law’,³³ the ECJ fashioned a constitutional basis for a ‘federal-type structure’ in Europe.³⁴ From this ‘federal type’ constitution, the ECJ asserted itself to be the final arbiter of what powers have and have not been conferred on the Union.³⁵

Under this ‘absolute’ conception of supremacy, Member State Kompetenz-Kompetenz has been criticized as an ‘anachronistic idea’ invoked under the ‘guise of protecting democracy’.³⁶ Judge Schiemann, for example, has reduced the defence of Member State sovereignty to ‘much the same instinctive defensive reactions as asking questions about a man’s virility’.³⁷According to scholars such as Habermas and Pernice, ‘National Courts are not authorised to monitor the limits of the transfer of national sovereign rights to the European level’.³⁸ This is so ‘even in the case of a conflict with the very substance of fundamental rights’ and even if EU law is ‘found to violate such fundamental rights or to be ultra vires’.³⁹

There is no explicit Treaty basis for this doctrine. It is based on two doctrinal justifications in ECJ jurisprudence – one pure constitutional and one normative.

The first (pure constitutional) justification holds, in essence, that the conferral of powers by the ‘peoples of Europe’ (Articles 1, 3(1) TEU) adds up to much the same thing as a single ‘people of Europe’, and the supremacy of EU law now derives from an autonomous source of legitimation that supersedes the national impulse to clutch back disputed territory. This can be seen in the ‘sovereignty building’ cases since the 1960s, wherein the ECJ justified supremacy by a direct connection between the peoples and the Union.⁴⁰ The European Parliament now provides a direct connection between a constituent people and EU law, not intermediated by national authorities.⁴¹ The supremacy of EU law is founded on ‘a common decision of the peoples of the Member States’ that cannot be questioned by national courts.⁴²

The second justification for supremacy is a normative one: the effective and uniform application of EU law.⁴³ This is most forcefully expressed when it is couched in terms of the rule of law,⁴⁴ legal certainty,⁴⁵ or the coherence of the EU legal order.⁴⁶ On this teleology, a failure to secure the uniformity and effectiveness of any EU law is an existential threat to the entire EU legal order as a whole.⁴⁷ This concern has animated ECJ jurisprudence since Internationale Handelsgesellschaft, where it held:

[T]he law stemming from the Treaty, an independent source of law, cannot [be] overridden by rules of national law, however framed […] without the legal basis of the Community itself being called into question.⁴⁸

1.1.2 The Federation of Sovereign States

In proclaiming autonomy and supremacy over all constitutional law, Europe’s judges enunciated a form of ‘federalism’. However, the inability to reconcile this with European constitutional theory meant, as Schütze writes, ‘In the absence of a federal theory beyond the State, European thought invented a new word – supranationalism – and proudly announced the European Union to be sui generis.’⁴⁹

Yet while this ‘sui generis’ claim pretended to reconcile two separate, sovereign constitutional orders, the hierarchy it enunciated was, in fact, a unitary monist legal order.⁵⁰ This was so because the ECJ ‘arrogated to itself the ultimate authority to draw the line between Community law and national law’.⁵¹ By denying the peoples of the Member States the final say over which powers they had or had not conferred on the Union, it denied the sovereignty of those peoples and in fact subjugated them under a unitary legal order.⁵² As the Italian Corte constituzionale noted, the ECJ ‘certainly considers that the source of legal norms of the Community and that of each Member State are founded on a single system’.⁵³

This led to irreconcilable tensions with persisting Member State sovereignty at the boundaries of EU law.

First, the declaration that the EU derived from its own autonomous Grundnorm didn’t simply deprive the Member States of their own. EU constitutionalism had not emerged from an act of a European people, but from the acts of public authorities – ‘governments, legislatures, courts(!)’.⁵⁴ Applying basic principles of constitutional theory, scholars found that it was ‘difficult – if not impossible to accept that “the founding treaties as well as each amendment agreed upon by the governments” appear as the direct expression of the corresponding will of the peoples of the Union’.⁵⁵ National constitutional courts agreed.⁵⁶ The EU was not a sovereign federal state,⁵⁷ but a federation of sovereign states (Staatenverbund) to which sovereign powers are delegated.⁵⁸ The German,⁵⁹ French,⁶⁰ Italian⁶¹ and Spanish⁶² constitutional courts all denied the autonomous ‘sovereignty’ claim of the European Parliament in their earliest encounters with it. The EU (including its parliament) was not founded by a pouvoir constituent originaire, but bound within competences set by international treaty.⁶³

Second, the institutions of the EU were not wholly supranational,⁶⁴ not wholly unknown to international law as claimed,⁶⁵ and those aspects which were supranational were not wholly democratic. The sole institution intended to embody a European people, the European Parliament, is bestowed with the weakest influence on the programme of legislation.⁶⁶ How could it be accepted that each EU norm is the direct expression of a European people, and yet, ‘the Community legislator does not receive any direct electoral mandate’?⁶⁷

Third, given the constitutional basis of conferral, ‘nearly all of the appellate courts balk at the claim of the ECJ that the European Treaties are the constitutions of an autonomous legal order’.⁶⁸ As will be shown, all twenty-eight Member State courts have come to assert that EU law takes effect not as an autonomous constitutionalism, but as a normative principle of national constitutional law. The Brunner (Germany) decision is perhaps the best known in that regard:

Germany is one of the ‘Masters of the Treaties’, which have established their adherence to the Union Treaty […] but could also ultimately revoke that adherence by a contrary act. The validity and application of European law in Germany depends on the application-of-law instruction of the Accession Act.⁶⁹

This assertion deprived autonomous European unitarism of its descriptive power because, as Maduro admits, ‘a different perspective is taken by national legal orders and national constitutions [requiring] a conception of the law which is no longer dependent upon a hierarchical construction’.⁷⁰

1.1.3 Constitutional Pluralism

Constitutional pluralism may now be said to have several strands, but the central tenet is that it departs from the Kelsenian emphasis on the locus of sovereignty in exchange for a normative conception of overlapping and interacting heterarchical (not hierarchical) claims.⁷¹ Constitutional pluralism accepts that neither authority – EU or Member State courts – can abandon the legal order they have been charged to protect.⁷² The benefit is that, in allowing theorists to ‘escape from the idea that all law must originate in a single power source’,⁷³ it ‘suggests that conflicts between the [ECJ] and national constitutional courts should be resolved through mutual accommodation rather than through uncompromising assertions of primacy’.⁷⁴

While not all can agree that pluralism justifies the competing claims of European and national constitutionalism, there are few who disagree that it describes them.⁷⁵ The virtue of constitutional pluralism lies in its ability to describe what courts will do, rather than what they should do as a matter of doctrinal principle.

In that respect, constitutional pluralism contains an inextricable (but oft-unacknowledged) thread of (Hartian) legal positivism.⁷⁶ This is so because not all constitutional disputes will be resolved through normative dialogue and, eventually, an irreconcilable conflict will arise.⁷⁷ Where it does, the methods constitutional pluralism has devised to resolve conflicts of law become little more than normative criteria for identifying which rule will in fact be recognized and applied in the positivist sense.

In that regard, the reality that matters for this book is that, whether one adopts a Kelsenian, normative or Hartian approach, Member States will often have the ‘final say’ as arbiters of the boundaries of EU law.⁷⁸ When applying MacCormick’s pluralist approach, ‘what matters […] is that a conflict rule must be valid from the vantage point of the norm taken as reference point of the legal system in order to be regarded as a rule of that legal system’.⁷⁹ On this approach, there are few jurists who would credibly argue that a declaration of invalidity by, say, the BVerfG with regard to the PSPP, or a European arrest warrant, would be ignored by German institutions, bound by the German constitution, for a normative claim by the CJEU that another rule should be applied.⁸⁰

This now seems accepted by Europe’s judges as an empirical matter, even if it is not admitted as a matter of doctrine. As Judge Maduro observes, while the doctrinal position is that EU law is the higher law, ‘National law still holds a veto power over EU law, and that is important even when it is not used’.⁸¹ Judge Lenaerts observes:

Day after day […] the [ECJ] must win the trust of Member States and national supreme courts as the ‘ultimate judicial umpire’ of [Union] competences […] The conceptual reason for this is rather straightforward: the Member States – and not the people as such – hold the Kompetenz-Kompetenz as makers of the constitution.⁸²

1.2 The Constitutional Boundaries of the EU Legal Order

1.2.1 Member State Kompetenz-Kompetenz

The first limit imposed by national constitutional orders on EU law is that of competence. Member States profess to retain for themselves the competence to decide on competences – the so-called Kompetenz-Kompetenz.⁸³ This is asserted in two ways: Through judicial ultra vires review by national courts (the judicial Kompetenz-Kompetenz), and through the act of treaty ratification itself (the so-called legislative Kompetenz-Kompetenz).⁸⁴

Such ultra vires review jurisdictions are based on intuitive logic: Under Articles 4(1), 5(1) and 5(2) TEU the limits of Union competence are governed by the principle of conferral, and under Articles 48(4) TEU, 49 TEU, 54 TEU and 357 TFEU, the EU acquires its competences when the Treaties are ‘ratified by the High Contracting parties in accordance with their respective constitutional requirements.’⁸⁵ This means that – supreme and legitimate within its bounds though it may be – there are nonetheless boundaries of the Union legal order beyond which the states are sovereign, and Member State constitutional law is the reference point for what those boundaries are.⁸⁶ Thus, Article 263 TFEU grants the CJEU jurisdiction to hear claims for lack of competence, but national courts have not-infrequently pointed-out that the same confederate foundations which constrain the EU legal order also apply to its court – the CJEU itself is a creature of the Treaties bound within its competences (and capable of acting ultra vires).⁸⁷ In Brunner (Germany), the BVerfG held:

[I]f European institutions or agencies were to treat or develop the Union Treaty in a way that was no longer covered by the Treaty in the form that is the basis for the Act of Accession […] German state organs would be prevented for constitutional reasons from applying them in Germany. Accordingly, the [BVerfG] will review legal instruments of European institutions and agencies to see whether they remain within the limits of the sovereign rights conferred on them or transgress them.⁸⁸

If that is so, the architects of EU fiscal federalism cannot rely on the authority of the ECJ alone to secure the good functioning of models that stretch the interpretation of EU competences, or depend on legal machineries placed beyond them.⁸⁹ As Irish Supreme Court Judge Charleton so puts it: ‘Cleary, the issue of what powers have been transferred remains a matter of German law for which only German courts have competency.’⁹⁰ The purpose of this Section 1.2.1 is therefore to evaluate this claim as a valid constitutional, normative and positivist description of the limits of EU law for the purposes of this book.

1.2.1.1 Pure Constitutional Evaluation of Member State Kompetenz-Kompetenz Adjudication

As a matter of pure constitutional law, the EU acquires its competences when the Treaties are ratified by the Member States in accordance with their respective constitutional requirements.⁹¹ The EU does not exist independently of the Treaties, and has no competences by right. The Union is ‘thus not “national” – that is: sovereign – in scope’.⁹² As the BVerfG has stated, ‘sovereignty under international law and public law requires independence from an external will precisely for its constitutional foundations’.⁹³ Other constitutional courts (including, at times, the ECJ)⁹⁴ arrive at similar evaluations of EU ‘sovereignty’.⁹⁵ As constitutional courts have been keen to assert, it is the Member States which are the ‘Masters of the Treaties’.⁹⁶

Without the limits of conferral, entering into the European Union would have been unconstitutional in all twenty-eight of Europe’s constitutional democracies (with one qualification)⁹⁷ reviewed in this chapter. The EU’s powers are carved-out from Member State constitutions and, nemo plus iuris, none of Europe’s constitutional democracies allow the disposition of the constitutional amending power by conferring Kompetenz-Kompetenz on the Union.⁹⁸ Maastricht (Spain) is characteristic:

[T]he Spanish parliament can grant or transfer the exercise of ‘powers derived from the Constitution’, but cannot dispense with the Constitution itself, contravening or permitting the contradiction of its provisions. The possibility of amending the Constitution is not a ‘power’ whose exercise can be granted.⁹⁹

In any event, Articles 48(4) TEU, 49 TEU, 54 TEU and 357 TFEU are quite clear on the manner of democratic legitimation for the acquisition of competence: the Treaties must be ratified by the Member States ‘in accordance with their respective constitutional requirements’. If supremacy is ‘founded on a common decision’ by a European people, then that ‘common decision’ was to resolve – by writing Articles 5 TEU, 48(4) TEU, 49 TEU, 54 TEU and 357 TFEU into the Treaties – that the EU cannot extend its own powers through any act not in accordance with Member State constitutional requirements. Thus, even if one accepts the pure constitutional justification for supremacy – that the conferral of powers by the ‘peoples of Europe’ adds up to much the same thing as a single constitutional ‘people of Europe’ – it remains that this legitimation can only ever flow within the limits of the EU’s conferred powers.¹⁰⁰ As asserted by the Spanish Tribunal Constitucional:

[T]he primacy set forth according to the Treaty […] is reduced expressly to the exercise of competences attributed to the European Union […] it is not a primacy with a general scope. […] Therefore, the primacy operates with regard to the competences transferred to the Union by the sovereign will of the State […] the competences whose exercise is transferred to the [EU] could not, without a breakdown of the Treaty itself, act as a foundation for the production of Community regulations whose content was contrary to the values, principles or fundamental rights of our Constitution.¹⁰¹

‘Absolute’ supremacy, however, implies something different. It implies that the Union may acquire competences other than in the manner provided by Articles 48(4), 49 and 54 TEU or 357 TFEU – that is, other than an act of ratification in accordance with constitutional law. This is because a well-meaning but erroneous ECJ intra vires ruling on an act outside EU law would effect a misappropriation of state power which nobody – neither the ‘peoples’ nor a ‘people of Europe’ – has voted to confer on the Union.¹⁰² Moreover, because supremacy applies within the scope of EU competence, the misappropriation of this ‘new’ EU competence permanently switches the power to determine law in that area from the Member State to the Union.¹⁰³ As the BVerfG warned in Weiss (Germany):

If the Member States were to completely refrain from conducting any kind of ultra vires review, they would grant EU organs exclusive authority over the Treaties even in cases where the EU adopts a legal interpretation that would essentially amount to a treaty amendment or an expansion of its competences.¹⁰⁴

For this reason, the German BVerfG has long held that it has an ultra vires review jurisdiction to decide whether the EU has stepped over the boundaries given to it.¹⁰⁵ According to that Court, an act of EU law that is manifestly outside the scope of competences, or an expansive interpretation of EU law that is ‘structurally significant’ to the allocation of competences in a manner ‘equivalent to an extension of the Treaty […] would not produce any binding effects for Germany’.¹⁰⁶

It is far from alone.

In Italy, the Corte constituzionale exercises Kompetenz-Kompetenz control over EU law under its ‘controlimiti’ (counter-limits) doctrine.¹⁰⁷ EU law is not autonomous, but is ‘founded upon […] Article 11(2) of the Constitution’.¹⁰⁸ It is only ‘within those areas in which the organs of the Community are competent’ that ‘the Community rule takes precedence’ over any rule of national law.¹⁰⁹

In France, the Conseil Constitutionnel exercises a priori constitutional control over acts of conferral,¹¹⁰ and the Conseil Constitutionnel,¹¹¹ Conseil d’État¹¹² and Cour de Cassation¹¹³ exercise a posteriori control of secondary law in excess of the constitutional authorization.¹¹⁴ Article 54 of the Constitution contains a nemo plus iuris rule,¹¹⁵ pursuant to which EU law cannot run counter to an ‘express contrary provision’ of the Constitution, unless the constituting power consents thereto.¹¹⁶

In Belgium the Cour constitutionnelle and Conseil d’État locate authority for the supremacy of EU law in Article 34 of the Belgian Constitution.¹¹⁷ This does not allow the disposition of Kompetenz-Kompetenz, and there is no basis for the application of EU law outside the national constitutional empowerment.¹¹⁸

In Denmark, an open-ended conferral, or the assumption of powers not specified in the Accession Act (including by judicial interpretation) would violate the Section 20 of the Constitutional act of Denmark.¹¹⁹ The Højesteret (Supreme Court) retains a powerful ultra vires jurisdiction: CJEU interpretations ‘must not result in the widening of the scope of Union powers’ and ‘it is for the Danish courts to decide whether EU acts exceed the limits for the surrender of sovereignty which has taken place by the Accession Act’.¹²⁰

In Sweden, the Högsta Domstolen (Supreme Court) and Högsta förvaltningsdomstolen (Supreme Administrative Court) derive authority for the direct effect,¹²¹ indirect effect¹²² and supremacy of EU law¹²³ from Chapter 10§6 of the Instrument of Government – not autonomous EU constitutionalism.¹²⁴ As Lebeck explains, a ‘legal act or decision from an EC/EU institution that exceeds the powers that have been delegated to the EC/EU would be ultra vires and hence not be valid law in the Swedish legal order’.¹²⁵

In the United Kingdom, the permissible scope of application of EU law was a function of the interpretation of the European Communities Act 1972,¹²⁶ and the UK courts retained an ultra vires jurisdiction to determine ‘whether the European Communities Act 1972 or any successor statute conferred any authority on the Court of Justice to exercise […] jurisdiction’ over issues outside the scope of authority so provided.¹²⁷

In Spain, the Tribunal Constitucional distinguishes between the primacía of EU law afforded by Section 93 of the Spanish Constitution (allowing EU law to supersede conflicting national law), and the supremacía of the Constitution itself (which both determines the status of EU law in the national order, and subjects it to integral constitutional guarantees).¹²⁸ In Constitutional Treaty (Spain) it held: ‘the primacy set forth for the Treaty and its resulting legislation […] is reduced expressly to the exercise of the competences attributed to the European Union […] by the sovereign will of the State’.¹²⁹

In Portugal, the supremacy of EU law derives from Articles 7(6) and 8(4) of the Portuguese Constitution, and the Tribunal Constitucional has held since ERDF (Portugal) that ‘there can be no exercise of the regulatory power without some basis in a lex anterior’.¹³⁰

In Greece, the Council of State (Συμβούλιο της Επικρατείας) locates constitutional authority for EU law in Article 28 of the Hellenic Constitution – not autonomous EU law.¹³¹ In Karella (Greece), the Council of State confirmed that the EU’s powers are constrained by the act of conferral, and the act of conferral is constrained by the Constitution.¹³²

In the Czech Republic, the Ústavní Soud (Constitutional Court) retains ultimate jurisdiction to determine ‘whether an act of the Union has exceeded the limits [of powers] which the Czech Republic transferred to the EU under Art. 10a of the Constitution’.¹³³

In Poland, Article 90(1) of the Constitution permits Poland to ‘delegate’ competences only ‘in relation to certain matters’, and the Trybunał Konstytucyjny (Constitutional Court) retains jurisdiction to ‘assess whether or not, in issuing particular legal provisions, the Community legislative organs acted within the delegated competences’.¹³⁴ Should they exceed them, ‘the principle of the precedence of Community law fails to apply with respect to such provisions’.¹³⁵

In Latvia, EU competences are legitimated by Article 68 of the Constitution, which allows Latvia to ‘delegate a part of its State institution competences’,¹³⁶ and the Satversmes tiesa (Constitutional Court) denies EU Kompetenz-Kompetenz.¹³⁷

In Lithuania, the Konstitucinis Teismas (Constitutional Court) denies EU Kompetenz-Kompetenz¹³⁸ and asserts its jurisdiction to ‘guarantee the supremacy of the constitution in the legal system as well as constitutional legality’ in the context of the EU.¹³⁹

In Ireland, Article 29.4.6 of the Constitution grants constitutional supremacy to EU law within the scope of the act of ratification, but the Supreme Court holds that this is only so provided that future expansions or amendments of EU law ‘did not alter the essential scope or objectives of the Union’.¹⁴⁰ Article 29.4.6 does not allow the state to dispose of Kompetenz-Kompetenz.¹⁴¹ EU law ‘has immunity but only if [the act of ratification] does not go outside the terms of the licence granted by [Article 29.4.6]’, failing which ‘such acts of the institutions of the Community as depend on [the act of ratification] for their status in domestic law would lose that status and would be of no effect in domestic law’.¹⁴²

The position is similar in Austria, where the Constitutional Court (VfGH) derives authorization for the supremacy of EU law from the Act on Accession of Austria (AAA), enacted by the ‘total revision’ procedure under Article 44(3) of the Austrian Constitution in 1993.¹⁴³ However, the 1993 ‘total revision’ does not cover future expansions of the EU legal order, and the EU does not have Kompetenz-Kompetenz.¹⁴⁴

In Estonia, the Constitution of Estonia Amendment Act (CEAA) provides that the Constitution will be applied ‘without prejudice to the rights and obligations arising from the Accession Treaty’.¹⁴⁵ However, as the Riigikohus (Supreme Court) has held, this applies only ‘within the spheres’ of EU competence,¹⁴⁶ and ‘does not authorise the integration process of the [EU] to be legitimised or the competences of Estonia to be delegated to the [EU] to an unlimited extent’.¹⁴⁷ A ‘more extensive delegation of the competence of Estonia to the European Union’ requires further consent from the Estonian people (by referendum).¹⁴⁸

In Romania, the Curtea Constituţională (Constitutional Court) distinguishes between the ‘priority’ or ‘precedence’ of EU law over legislation and the ‘supremacy’ of the Constitution (which both determines the effect of EU law and prevails over it).¹⁴⁹ This precludes EU Kompetenz-Kompetenz. EU law is derived from Article 148 of the Constitution (which permits ‘exercising’ ‘certain powers’ with other states), and EU acts are ‘norms interposed within the constitutionality control’.¹⁵⁰

In Bulgaria, the Constitutional Court (Конституционен съд) holds that the supremacy of EU law is justified because ‘the institutions of the European Communities act within their competences [which are] subject to ratification’.¹⁵¹ Authorization for conferral is ‘not unlimited’,¹⁵² and the EU can only acquire powers through acts of the people, ‘at their own will, through the National Assembly elected by them’.¹⁵³

In Slovenia, the Ustavno Sodišče (Constitutional Court) holds that EU law becomes ‘internal constitutional principles that have the same binding effect as the Constitution’ by virtue of Art. 3a of the Slovenian Constitution – it is neither autonomous nor constitutionally supreme.¹⁵⁴ In SNHCA (Slovenia), the Court declined to endorse EU supremacy over the Constitution and described Kompetenz-Kompetenz as a permanent constraint on conferral.¹⁵⁵

In Slovakia, the Ústavný Súd (Constitutional Court) holds that EU law has the status of international treaties under Article 7(2)¹⁵⁶ or 7(5)¹⁵⁷ of the Constitution (or both), meaning EU law does not have Kompetenz-Kompetenz and is subject to the Constitution.¹⁵⁸ In Constitutional Treaty (Slovakia), the Court denied the EU was a ‘state union’ with Kompetenz-Kompetenz.¹⁵⁹

In Finland, constitutional review is exercised by the Perustuslakivaliokunnan (Constitutional Committee), which holds that neither the Act of Accession nor Sections 94–95 of the Constitution (the bases for conferral and application of EU law) can endanger the democratic foundations of the Constitution, in particular Kompetenz-Kompetenz.¹⁶⁰ As Ojanen observes, ‘the Committee’s message is that Kompetenz-Kompetenz remains – and should continue to remain, in the hands of Finland’.¹⁶¹

The Constitution of Malta states that Parliamentary legislation made in conformity with international/EU obligations are ‘Subject to the provisions of [the Maltese] Constitution’,¹⁶² and an amendment to Malta’s constitutional supremacy clause (Article 6) could not be achieved.¹⁶³ Thus, as Xuereb explains, because the authority for EU law must ‘take the form of an Act of Parliament passed in virtue of the Constitution’, the Constitutional Court (Qorti Kostituzzjonali) retains the ‘final say’ on the scope and effect of EU law within the constitutional system.¹⁶⁴

In Croatia, the supremacy of EU law is derived from Articles 141–143 and 145 of the Croatian Constitution, not autonomous EU constitutionalism.¹⁶⁵ Under Article 2 of the Constitution, Croatia ‘retain[s] its sovereign right to decide upon the powers to be so delegated’,¹⁶⁶ and the Ustavni sud (Constitutional Court) holds that ‘the Constitution is, by its legal nature, supreme to EU law’.¹⁶⁷

In Hungary, the Magyarország Alkotmánybírósága (Constitutional Court) has asserted judicial Kompetenz-Kompetenz under the sovereignty provisions of both the 1989 Constitution and the 2011 Fundamental Law.¹⁶⁸

In Cyprus, the Ανώτατο Δικαστήριο (Supreme Court) derives the supremacy of EU law from Article 1 A of the Constitution,¹⁶⁹ a constitutional exceptive clause introduced in 2006 after the Supreme Court ruled EU law could not prevail over conflicting Constitutional provisions.¹⁷⁰ However, this does not confer Kompetenz-Kompetenz. Under the Cypriot Constitution, treaties ‘shall only be operative and binding on the Republic when approved by a law made by the House of Representatives’,¹⁷¹ and the Constitution remains ‘the supreme law of the Republic’.¹⁷² Consequently, ‘any delegation of transfer of competences is understood as an expression of the will of the sovereign state and as a matter of choice’.¹⁷³

In Luxembourg and the Netherlands, national courts are prohibited from reviewing the constitutionality of acts ratifying international treaties, and international law prevails over constitutional law.¹⁷⁴ This has led to a debate over whether EU law would apply even if the constitutional bases for conferral were abolished – idem est, ‘the Dutch constitution is entirely irrelevant in that regard’.¹⁷⁵ However, this would seem to be overstated. In both countries the constitutional supremacy of EU law cannot arise unless ratified by special majorities in accordance with the constitution.¹⁷⁶ Besselink, Claes and De Witte point out that the early decisions of the Hoge Raad¹⁷⁷ and Raad van State¹⁷⁸ derived authority to disapply national law from the Dutch Constitution, and there is little to have altered this position.¹⁷⁹ The position is similar in Luxembourg, where the special ratification procedure in Article 114(2) of the Constitution was necessary to ensure the constitutionality of the Maastricht Treaty,¹⁸⁰ and ‘the [Luxembourg] Chambre is clearly the holder of revision power’.¹⁸¹

In sum, as a matter of pure constitutional law, no Member State accepts the absolute supremacy of EU law over the Kompetenz-Kompetenz. In all Member States, EU acts not conferred in accordance with the constitution are, in principle, invalid in the national legal order without (at minimum) parliamentary ratification or constitutional amendment. As the BVerfG concludes:

The ‘Constitution of Europe’, international treaty law or primary law, remains a derived fundamental order […] according to the principle of conferral, without the possibility for the European Union of taking possession of Kompetenz-Kompetenz.¹⁸²

1.2.1.2 Normative Evaluation of Member State Kompetenz-Kompetenz Adjudication

This section evaluates competing normative claims over Kompetenz-Kompetenz adjudication. This is necessary because ‘absolute’ EU supremacy over Kompetenz-Kompetenz adjudication also relies upon a normative claim: That is, even if national courts retain formal authority over the status of EU law under constitutional law, they must accept that the ‘effectiveness and uniformity of EU law’ is of such normative importance that the constitutional authorization for EU law will always outweigh any conflicting constitutional norms – even those which constrain the act of ratification.¹⁸³ Take, for example, the apocryphal statement of EU supremacy by Pernice:

A residual control of the Court of Justice by national Constitutional courts in cases of continuous and evident violations of fundamental rights or [ultra vires acts] as an element of balance of powers is excluded, since […] non-application of Community law in one Member State would jeopardize the status of legal equality of the Union citizens which is the foundation of its functioning.¹⁸⁴

Suffice it here to state that this normative claim is not accepted in any of the constitutional courts catalogued in this book.¹⁸⁵ When the Union acquires its competences upon ratification by the Member States ‘in accordance with their respective constitutional requirements’,¹⁸⁶ the supremacy of EU law is secured within the constitutional order because conferral cannot be done in such a way that it would violate or vitiate conflicting norms in the constitution. As the Spanish Tribunal Constitucional so puts it, ‘public authorities are no less subject to the Constitution when they act in the international or supranational relations than when they exercise their competences ad intra’.¹⁸⁷ The principle that emerges here is that national courts cannot hold the ‘effectiveness and uniformity’ of EU law over the constitutional boundaries of conferral, because the EU is a derived legal order circumscribed by constitutional norms exerted on conferral itself.¹⁸⁸ From this common foundation, Member States evince three approaches to situating the normative supremacy of EU law within the constitution.¹⁸⁹

In a first group of countries, consisting of France (to 2006),¹⁹⁰ Denmark,¹⁹¹ Greece,¹⁹² Spain,¹⁹³ the Czech Republic,¹⁹⁴ Poland,¹⁹⁵ Slovenia,¹⁹⁶ Slovakia,¹⁹⁷ Romania,¹⁹⁸ Bulgaria,¹⁹⁹ Latvia,²⁰⁰ Malta,²⁰¹ Croatia²⁰² and Lithuania,²⁰³ the constitutional basis for EU law is subject to a nemo plus iuris rule which prevents the state from conferring the competence to exercise its powers in a manner contrary to the constitution. Provisions of the constitution in conflict with the treaty must be amended and, if they cannot be so amended, the treaty (or the application thereof) will be unconstitutional.²⁰⁴

In a second group of countries, consisting of Germany,²⁰⁵ Italy,²⁰⁶ France (from 2006),²⁰⁷ the UK,²⁰⁸ Ireland,²⁰⁹ Portugal,²¹⁰ Austria,²¹¹ Sweden,²¹² Estonia,²¹³ Finland,²¹⁴ Belgium²¹⁵ and Hungary,²¹⁶ the constitutional empowerment for EU law does apply irrespective of conflicting constitutional law, either by derogation or by an extraordinary instrument that bestows heightened rank on EU law. However, EU law does not take effect autonomously, and the derogation does not apply to important constitutional principles which are either beyond the reach of the legislator, or anyways always of greater normative weight than the effectiveness of EU law. This model includes, for example, British parliamentary sovereignty,²¹⁷ the Italian controlimiti doctrine²¹⁸ and the German ‘eternity clause’.²¹⁹ This model can be seen at work in such cases as Grogan (Ireland),²²⁰ or Taricco II (Italy),²²¹ where normatively important constitutional principles trounced the imperative of the ‘effectiveness and uniformity’ of EU law.

In a third group of countries, consisting of the Netherlands, Luxembourg and Cyprus, EU law is normatively supreme over the constitution because judicial review of the EU Treaties is precluded by the constitution. However, even then it seems EU law is not normatively supreme over democracy: The EU can have no powers without a legislative act of conferral made in accordance with the constitution.²²²

Whatever group they fall into, all of these jurisdictions have two features in common. First, no Member State accepts that the ‘uniformity and effectiveness’ of EU law is of such normative importance that it prevails over constitutional control of Kompetenz-Kompetenz. As Member State courts have been keen to point out, the ‘effectiveness and uniformity’ of EU law within its competences cannot depend on the appropriation of national powers outside them.²²³ For this reason, according to the BVerfG, ultra vires review does not ‘factually contradict’ supremacy,²²⁴ and ‘a substantial risk to the uniform application of [EU] law does not result’.²²⁵

The second thing they have in common is that acts of conferral are made of the same fabric as the constitution from which they have been cut – they can have no ‘extra-constitutional’ properties other than those ascribed by the constitution.²²⁶ In virtually all Member States, a conflict between EU law and the constitution from whence it has been carved is either ‘infra-constitutional’ – that is, the EU law is not of constitutional rank at all;²²⁷ or ‘intra-constitutional’ – a clash between two national constitutional provisions: the one authorizing EU law and whatever one is in conflict with it.²²⁸ The EU provision is given a higher or lower normative weight in a conflict depending on which country and which values are concerned, but in all instances the consequence of EU law spilling over into conflict with another constitutional provision is, as the Spanish Court puts it, ‘a fact which must be considered as established from the perspective of [national] law’,²²⁹ and a matter of ‘the selection of the rule to be applied’.²³⁰

1.2.1.3 Positivist Evaluation of Member State Kompetenz-Kompetenz Adjudication

This brings us to a positivist consideration of Kompetenz-Kompetenz adjudication. If the absolute supremacy of EU acts, as interpreted by the ECJ, is to be accepted as the rule of recognition for identifying which models of fiscal federalism are implementable in the EU, it must in fact provide an authoritative and reliable account of what is and is not safe constitutional ground to install legal instruments of public economics. This is particularly so when dealing with such things as temperamental bond markets and the politico-economic incentives of restive electorates. Certainty, expansive intra vires rulings of the ECJ over such instruments cannot be so constitutionally fraught that they risk destabilizing the entire fiscal architecture each time they are issued.

In that regard it must be recalled that, in all Member States, the constitutional authorization for the application of EU law is a legislative instrument enacted under a specific constitutional window.²³¹ Debates about whether it is legitimate for national courts to conduct ultra vires review are, first and foremost, debates about national constitutions.²³² Given this is so, a coercive approach to imposing supremacy in areas considered outside the boundaries of conferral is, with certainty, counterproductive to the goal of effectiveness and uniformity in the EU legal order.²³³ As Kumm notes, ‘The likelihood that all laws will in fact be applied throughout the community will decrease as the probability that a particular law will be struck down on constitutional grounds by a national court increases.’²³⁴ Judge Maduro concurs:

A hierarchical alternative imposing a monist authority of European law and its judicial institutions over national law would be difficult to impose in practical terms and could undermine the legitimacy basis on which European law has developed.²³⁵

Such admissions align with statements of law from the Member States – ‘absolute’ supremacy cannot be applied as the rule governing the validity of contested acts without jeopardizing the integrity of the Union itself.²³⁶ The BVerfG states:

[I]t is not enough simply to speak of the ‘precedence’ of Community law over national constitutional law in order to justify the conclusion that Community law must always prevail over national constitutional law because, otherwise, the Community would be put in question.²³⁷

For the architects of fiscal federalism, it would be foolish to proceed on the cheerful basis that Member States daren’t apply the jurisdictions they have set out, just to preserve the good functioning of some ideal model that impinges the boundaries of competence. Attempts to assert ‘absolute’ supremacy over Kompetenz-Kompetenz adjudication have provoked several of these jurisdictions – with immediate and deleterious effects on the uniformity and effectiveness of EU law.

Perhaps most recently, in Ajos (Denmark), the Højesteret refused to disapply national employment legislation as directed by the ECJ in Ajos (CJEU),²³⁸ holding that ECJ case law on age discrimination was itself ultra vires the Danish act of accession.²³⁹ In Gauweiler (Germany) the BVerfG inveighed against a permissive interpretation of ECB competence by the ECJ and placed six conditions on the operation of a (technically supreme) EU law bond-buying programme.²⁴⁰ In Weiss (Germany) the BVerfG held that the ECJ’s permissive interpretation of the same competence in Weiss (CJEU) ‘manifestly exceeded the judicial mandate conferred upon the CJEU in Art. 19(1) TEU’ such that ‘the CJEU Judgment itself constitutes an ultra vires act and thus has no binding effect [in Germany]’.²⁴¹

Even where such outright conflicts are avoided through subtler shades of interpretive disobedience, the jurisprudence cited in this chapter is replete with examples of EU law bending around constitutional guarantees at the margins of competence.²⁴² As the BVerfG observed in R v. Oberlandsgeright (Germany) (citing 27 judgments from ten countries):

The overwhelming majority of the constitutional and supreme courts of other Member States shares for their respective sectors in the view of the [BVerfG] that the (application) primacy of Union law is not unlimited, but that are drawn to it by the national (constitutional) limits.²⁴³

Legal scholars trawling the case law make similar observations. Woods and Watson find that ‘all the constitutional courts of the Member States regard themselves as having the power to review the boundary of EU competence’.²⁴⁴ Surveys by Grabenwarter,²⁴⁵ Claes,²⁴⁶ Kumm and others reach similar conclusions: ‘National Constitutional Supremacy is a legal rule that governs practice as a matter of fact, and that is all there is to it.’²⁴⁷

1.2.2 Member State Constitutional Identity

The second constitutional boundary imposed on the EU legal order is an absolute one: Not only have some powers not been conferred on the EU, but some constitutional powers or principles can never be transferred to the Union or vitiated by conflicting EU law. These are typically referred to as the limits of ‘constitutional identity’ – inalienable, inviolable structures or principles so integral to the constitutional order that they either cannot be formally altered by the amending power at all; or otherwise impose material constraints that cannot be released without effecting a ‘total revision’ or legal revolution that would result in a different constitutional system – a different constitutional identity.²⁴⁸ Constitutional identity principles ensure that amendments and evolutions of constitutional law remain within the framework of the constitution and consistent with its foundational principles. They may often be recognized apart from ‘ordinary’ constitutional principles by their various functions: preserving popular or state sovereignty,²⁴⁹ safeguarding the distinction between constitution-making and constitution-amending authority²⁵⁰ and setting limits on the disposal of state competences and the supremacy of EU law.²⁵¹

The unamendable ‘eternity clause’ in the 1949 German Basic Law is the most notorious in this respect, but many other constitutional courts and committees have also asserted some ‘inviolable core’ integral to the constitution. The Belgian Cour constitutionnelle,²⁵² the Bulgarian Конституционен съд (Constitutional Court),²⁵³ the Croatian Ustavni Sud,²⁵⁴ the Czech Ústavní Soud,²⁵⁵ the Danish Højesteret,²⁵⁶ the Estonian Riigikohus,²⁵⁷ the Finnish Perustuslakivaliokunnan,²⁵⁸ the Austrian Verfassungsgerichtshof,²⁵⁹ the French Conseil Constitutionnel,²⁶⁰ the Greek Συμβούλιο της Επικρατείας (Council of State),²⁶¹ the Hungarian Magyarország Alkotmánybírósága,²⁶² the Irish Supreme Court,²⁶³ the Italian Corte constituzionale,²⁶⁴ the Latvian Satversmes Tiesa,²⁶⁵ the Lithuanian Konstitucinis Teismas,²⁶⁶ the Polish Trybunał Konstytucyjny,²⁶⁷ the Portuguese Tribunal Constitucional,²⁶⁸ the Romanian Curtea Constituţională,²⁶⁹ the Slovak Ústavný Súd,²⁷⁰ the Slovenian Ustavno Sodišče,²⁷¹ the Spanish Tribunal Constitucional,²⁷² the Swedish Konstitutionsutskottet,²⁷³ the UK Supreme Court²⁷⁴ and the German BVerfG²⁷⁵ have all asserted that some constitutional powers or principles cannot be disposed-of under the national constitution or vitiated by conflicting EU law, either de lege lata or at all.²⁷⁶

The 2010 Polish Tribunal Konstytucyjny encapsulates the jurisprudence thusly:

Constitutional identity is a concept which determines the scope of excluding – from the competence to confer competences – the matters which constitute […] ‘the heart of the matter’, i.e., are fundamental to the political system of a given state.²⁷⁷

For the architects of European fiscal federalism, this presents a dilemma. This is so because, under Article 4(2) TEU, the Union itself is under a duty to ‘respect the national identities of its Member States’, and the ECJ disavows the interpretation that this allows constitutional identities to limit the scope of EU law.²⁷⁸ As Judge Lenaerts has written, ‘There simply is no nucleus of sovereignty that the Member States can invoke, as such, against the Community’ – even when the Treaty ‘expressly acknowledges the existence of residual powers for the Member States’.²⁷⁹ Instead, the ECJ has interpreted national identity under Article 4(2) as encompassing an open-ended list of cultural, social or legal values not common enough to be ‘general principles’ on their own right into a single principle that Lenaerts calls ‘value diversity’ – over which the CJEU then has jurisdiction.²⁸⁰ ‘National identity’ includes ‘constitutional identity’.²⁸¹ In the eyes of EU law, national identity is no different than other ‘legitimate aims’ whose purpose is, as stated in Cassis, ‘not to reserve certain matters to the exclusive jurisdiction of the Member States’ but to restrict derogations to the extent justified against the objectives of EU law.²⁸² On this reading, Article 4(2) TEU does not brace the containment walls of EU competence – it subsumes those boundaries within the EU legal order and gives the ECJ jurisdiction to examine their merit.

Member State constitutional identity and CJEU national identity jurisdictions therefore profess to govern the same thing, but draw very different red lines around the contours of EU competence. What the architects of fiscal federalism must determine is whether – as the ECJ maintains – it is the sole and final arbiter of what is or is not an infringement of constitutional identity, capable of ‘ousting’ the jurisdictions of national courts;²⁸³ or whether it is national courts that will determine what the ultimate boundaries of the EU legal order (and EU fiscal federalism) will be.

1.2.2.1 Pure Constitutional Evaluation of Constitutional Identity Review

The first task must be to compare the pure constitutional authority for these jurisdictions. The constitutional basis for the CJEU’s national identity jurisdiction is Article 4(2) TEU – a provision of EU law. Introduced at Maastricht to reassert ‘that the external limit on the exercise of the Union’s conferred powers are the fundamental constitutional structures of the Member States’,²⁸⁴ it reads:

The Union shall respect the equality of Member States before the Treaties as well as their national identities, inherent in their fundamental structures, political and constitutional, inclusive of regional and local self-government.²⁸⁵

Member State constitutional identity jurisdictions, by contrast, derive from constitutional provisions and principles which bind acts of conferral outside the EU legal order before they become EU law, and therefore define what may never be conferred on the Union and is therefore outside the EU legal order altogether.²⁸⁶

In Germany, the precedence of EU law is ‘limited by the Basic Law’s constitutional identity that, according to Art. 23(1) in conjunction with Art. 79(3) [BL] is neither open to constitutional amendments nor to European integration’.²⁸⁷ Those articles, the so-called ‘eternity clause’ (Article 79(3) BL) and the constitutional safeguard clause (Article 23(1) BL), entrench the highest principles of the German state from constitutional change by amendment or conferral.²⁸⁸

In Italy, the Corte constituzionale has held since 1973 that ‘fundamental principles of the Italian Constitution’ impose controlimiti (counter-limits) to the entry of EU law, and that the Italian Court would ‘always control the continuing compatibility of the Treaty with fundamental principles’.²⁸⁹ A violation of these principles by EU law will result in its invalidity,²⁹⁰ or, if the Treaty is itself is in conflict with the Constitution, ‘the radical and disruptive remedy of withdrawal from the European Union’.²⁹¹

In France, the ‘identité constitutionnelle de la France’ is assimilated to the ‘conditions essentielles d’exercise de la souveraineté’,²⁹² and the ‘structures constitutionnelles’ of the indivisible, secular, democratic and social Republic.²⁹³ The ‘ordinary’ supremacy of EU law is derived from Articles 55, 88–1 and 88–2 of the Constitution, but the French Constitution remains ‘at the pinnacle of the national legal order’ and does not permit ratification of EU law that ‘calls into question the rights and freedoms guaranteed by the Constitution or runs contrary to the essential conditions for the exercise of national sovereignty’.²⁹⁴

In Spain, the Tribunal Constitucional recognizes an ‘essential nucleus of powers’,²⁹⁵ which impose ‘material limits imposed on the transfer [to the EU] itself’.²⁹⁶ Said material limits are understood as ‘the sovereignty of the State, or our basic constitutional structures and of the system of fundamental principles and values set forth in our Constitution, [including] fundamental rights’.²⁹⁷

In Denmark, the Højesteret has held since Carlsen (Denmark) that ‘no transfer of powers can take place to such an extent that Denmark can no longer be considered an independent state’ or undermine the ‘democratic system of government’,²⁹⁸ and the Højesteret rejects the claim that EU supremacy ousts the ‘Danish court’s testing of the constitutionality of acts and EU Acts’.²⁹⁹

In the United Kingdom, constitutional identity inheres in the doctrines of parliamentary sovereignty,³⁰⁰ the rule of law, legality and constitutional statutes.³⁰¹ The supremacy of EU law was effected by treating the European Communities Act 1972 as one such constitutional statute, but parliamentary sovereignty is ‘fundamental to the United Kingdom’s constitutional arrangements’ and ‘EU law can only enjoy a status in domestic law which that principle allows’.³⁰² ECJ rulings were not to be interpreted so as to ‘question the identity of the national constitutional order’,³⁰³ or exert jurisdiction over ‘issues integral to the identity of the nation state’.³⁰⁴

In Portugal, the fundamental principles of the Portuguese Constitution are entrenched by an unamendability clause (Article 288), and Articles 7(6) and 8(4) of the Constitution contain constitutional safeguard clauses that condition EU law on ‘respect for the fundamental principles of a democratic state based on the rule of law’.³⁰⁵ The Tribunal Constitucional ‘has never accepted the supremacy of EU law over the Constitution’, and the prevailing view is that Articles 7(6) and 8(4) provide a basis to review EU law against its fundamental principles.³⁰⁶ Most recently, in its conditionality case law, the Tribunal Constitutional identified a ‘hard core’ of the rule of law and annulled measures it considered were ‘binding on the Portuguese State [as] legal instruments [of] European Union law’,³⁰⁷ holding:

[B]inding or not […] in a multilevel Constitutional system, in which several legal orders interact, internal legal norms cannot breach the Constitution […] European Union law itself establishes that the Union respects the national identity of its Member States, reflected in the fundamental political and constitutional structures of each of them.

(see Article 4(2) TEU)³⁰⁸

In Belgium, the Cour constitutionnelle holds that the Constitution ‘does not allow a discriminating derogation to the national identity inherent in the fundamental structures, political and constitutional, or to the basic values of protection offered by the Constitution’.³⁰⁹ As Gérard and Verrijdt encapsulate: the Court ‘forbids attributions of powers to the EU, and the application thereof by the EU organs, insofar as they encroach upon Belgian national identity or the basic values of constitutional rights protection’.³¹⁰

In the Czech Republic, the Ústavní Soud has held since Sugar Quotas that ‘the essential attributes of a democratic state governed by the rule of law […] remain beyond the reach of the Constituent Assembly itself’ and that ‘should developments in the EC, or the EU, threaten the very essence of state sovereignty of the Czech Republic or the essential attributes of a democratic state governed by the rule of law, it will be necessary to insist that these powers be once again taken up by the Czech Republic’s state bodies’.³¹¹ In the event of a lesser but clear conflict with EU law, ‘the constitutional order of the Czech Republic, in particular, its material core, must take precedence’.³¹²

In Latvia, the Satversmes tiesa holds that ‘National identity of the Member States is an essential basis of the EU’,³¹³ and the fundamental principles of the Satversme (Constitution) place an ultimate stop on the conferral and application of EU law – including the duty of conforming interpretation.³¹⁴ The fundamental principles of the Satversme ‘cannot be infringed by introducing amendments to the Satversme’ and delegation of competencies to the EU ‘cannot exceed the rule of law and the basis of an independent, sovereign and democratic republic based on the basic rights’.³¹⁵

In Lithuania, Article 1 of the Constitutional Act on Membership (Lithuania) allows Lithuania to ‘share with or entrust’ state competences, but only with a Union that ‘respects the national identity and constitutional traditions of its Member States’.³¹⁶ As interpreted by the Konstitucinis Teismas, the Lithuanian constitutional identity comprises the independent democratic republic, encompassing the independence of the state, democracy, the republic, innate human rights and freedoms,³¹⁷ and the supremacy of the constitution over EU law itself.³¹⁸

In Romania, the Curtea Constituţională holds that the ‘supreme values’ of the Constitution entrenched by its unamendability clause (Article 152), in particular the rule of law and the supremacy of the Constitution, impose permanent constraints on the supremacy of EU law.³¹⁹

In Croatia, the Ustavni sud holds that ‘the Constitution is, by its legal nature, supreme to EU law’,³²⁰ and amendments to the constitution by referendum cannot alter ‘the structural characteristics of the Croatian constitutional state, or in other words, of its constitutional identity, including the highest values of the constitutional order of the Republic of Croatia (Article 1 and Article 3 of the Constitution)’.³²¹

The Constitutional Court of Bulgaria (Конституционен съд) holds that EU law becomes part of Bulgarian law only in so far as it is ‘in compliance with the provided conditions’ of the Constitution, and that Bulgaria’s ‘constitutional identity is preserved’ in participation in the EU.³²² Constitutional identity finds locus in Article 158 of the Constitution, the onerous ‘Grand National Chamber’ revision procedure that entrenches ‘the people’s sovereignty, supremacy of the Constitution, political pluralism, separation of powers, rule of law and judicial independence’ from amendment.³²³ In particular, under the ‘democratic constitutional model’ of Bulgaria, the National Assembly must retain the ‘basic powers […] assigned by the Constitution’.³²⁴

In Slovenia, EU law is given an equivalent (but not superior) rank to the Constitution,³²⁵ and both sovereignty and the pravna drźava (state governed by the rule of law) limit the permissible transfer of powers under Article 3a of the Constitution (the basis for conferral).³²⁶

In Greece, acts of conferral are constrained by both a constitutional safeguard clause (Article 28(3)) and an unamendability clause (Article 110(1)) which place the Parliamentary Republic, the powers of the state and basic civil and political rights beyond amendment or conferral.³²⁷ The Council of State has sometimes been at pains to interpret the Hellenic Constitution in conformity with EU law,³²⁸ but it has also formally denied the supremacy of EU law over it.³²⁹ In DI.KATSA (Greece) the Council of State resolved a conflict with EU law in favour of the Constitution, concluding that it was ‘clearly necessary for the preservation of the national identity’,³³⁰ and in Jus Soli (Greece) it asserted that Article 4(2) TEU guaranteed respect for national identity in Article 1(3) of the Constitution (as interpreted by the Council of State).³³¹

In Sweden, Chapter 10§6 of the Instrument of Government states that conferral must not affect the Basic Principles of the Form of Government, and that EU membership is presupposed on an equivalent level of fundamental rights protection to the Swedish Constitution and the ECHR.³³² The clause was modelled after Germany’s ‘constitutional identity’ jurisprudence,³³³ and ‘implies a serious reservation against the principle of supremacy’.³³⁴ According to the Constitution Committee (Konstitutionsutskottet), law-making powers conferred on the EU cannot modify fundamental principles of Sweden’s constitutional system.³³⁵ The Högsta Domstolen has not openly invalidated EU law on this basis, but it has treated national implementations of EU law as purely internal law and interpreted them in conformity with basic principles, even though this has appeared prima facie contrary to EU law.³³⁶

In Poland, the Trybunał Konstytucyjny has long asserted an ‘untouchable material core’ inherent in the Polish constitutional identity.³³⁷ In Lisbon (Poland), it held:

The Constitutional Tribunal shares the view expressed in the doctrine that the competences, under the prohibition of conferral, manifest about a constitutional identity […] the following should be included among the matters under the complete prohibition of conferral: decisions specifying the fundamental principles of the Constitution and decisions concerning the rights of the individual which determine the identity of the state, including, in particular […] human dignity and constitutional rights […] statehood […] democratic governance […] the rule [of] law […] social justice […] and the prohibition to confer the power to amend the Constitution and the competence to determine competences.³³⁸

Although long asserted against external predations by international law, since 2015 executive reforms aimed at undermining the Polish judiciary under the guise of a political ‘constitutional identity’ narrative have instead appeared to work ‘in violation of clear constitutional standards and in conflict with their interpretation laid down in the case law of the Constitutional Tribunal, the Supreme Court and the legal doctrine that has developed since the adoption of the 1997 Constitution of the Republic of Poland’.³³⁹

In Hungary, the Magyarország Alkotmánybírósága enunciated a material form of constitutional identity constraint under the ‘European clauses’ of both the 1989 Constitution (Article 2/A)³⁴⁰ and the 2011 Fundamental Law (Article E (2)).³⁴¹ However, the absence of any procedural entrenchment provisions in the Constitution (all constitutional provisions are formally amendable with the same 2/3 majority) has left it comparatively defenceless against internal predations by the Hungarian executive.³⁴²

In Ireland, the Supreme Court holds that Article 29.4.6 of the Constitution (the basis for conferral) does not bestow a power on state institutions to dispose of their own competences or ‘qualify, curtail or inhibit the existing sovereign power’,³⁴³ and in Grogan (Ireland), the Court rejected the supremacy of EU law over fundamental constitutional guarantees.³⁴⁴ As Cahill concludes: The Supreme Court will ‘defend the Irish constitutional legal order on almost exactly the same terms as the constitutional courts in other Member States’.³⁴⁵

In Estonia, the CEAA grants supremacy over Estonian law,³⁴⁶ including constitutional law,³⁴⁷ but this is only so ‘provided that the fundamental principles of [the Constitution] are respected’.³⁴⁸

In Austria, the Basic Principles of the Austrian Constitution (democracy, the Republic, the federal state and the rule of law) ‘are considered to form a constitutional core that may not be limited by EU law’ and ‘present limitations to European integration’.³⁴⁹ If EU law is in conflict with Basic Principles, the VfGH ‘has to […] declare that the relevant rules of EU law are not applicable in Austria […] [t]hey have to be regarded as void acts’.³⁵⁰ ESM (Austria) and TSCG (Austria) indicate that a conferral of economic competences on EU institutions would contradict the Basic Principles.³⁵¹

In Slovakia, Articles 7(2) and 7(5) of the Constitution grant EU supremacy over ‘laws’,³⁵² but amendments to grant supremacy over ‘constitutional law’ or the ‘transfer a part of the exercise of its sovereignty’ were rejected.³⁵³ The Ústavný Súd has held that Slovakia ‘can only enter into a state union […] in which there is no violation of [the Constitution], in particular Art. 1 [the sovereign democratic state governed by the rule of law]’,³⁵⁴ and that ‘the referential framework of constitutional review remains limited to the norms of the Slovak constitutional order’ even after accession to the EU.³⁵⁵As Kovacs concludes, it is clear that Court ‘has the power to review EU law if this is indispensable to protect the constitutional identity of the country’.³⁵⁶

In Finland, the Perustuslakivaliokunnan holds that EU law cannot affect the democratic foundations of the sovereign republic under Section 1 of the Constitution,³⁵⁷ and Section 94(3) now contains a constitutional safeguard clause that constitutionalizes this interpretation.³⁵⁸ Under this constraint, the Perustuslakivaliokunnan has found that Finland cannot confer its competence for controlling financial liabilities on an international body voting by QMV,³⁵⁹ and EU law cannot weaken domestic standards of fundamental rights.³⁶⁰

From this tour it is clear that, as a matter of pure constitutional law, the CJEU’s ‘national identity’ jurisdiction under Article 4(2) TEU does not, and could not, grant jurisdiction over the grounds for constitutional identity review that bind acts of conferral in any of these countries. For Member State constitutional courts, therefore, Article 4(2) TEU is merely ratificatory of ‘the thrust of the jurisprudence of numerous domestic constitutional courts on the relationship between EU law and national constitutional law’.³⁶¹ The Spanish Tribunal Constitucional, for example, has stated that ‘the limits referred to by the reservations of said constitutional justifications now appear proclaimed unmistakeably by the Treaty’.³⁶² Numerous other courts have cited Article 4(2) TEU as though it was merely ratificatory of their own constitutional identity jurisdictions.³⁶³

This is so because the interpretation of constitutional identity involves the interpretation of national constitutional laws, and the CJEU lacks jurisdiction to do so under Article 19 TEU.³⁶⁴ As a matter of law, it is blind to the ‘identities’ which it professes to define respect for. Indeed, the ECJ has itself accepted (though not always)³⁶⁵ that only national courts can define what comprises national identity.³⁶⁶ Yet, as Preshova points out, this is not enough: it remains that when deciding the weight of such claims in a conflict with EU law, the ECJ will still ‘enter into a forbidden zone of determining the content and scope of the constitutional identity of a Member State. This is in essence contrary to Article 19 TFEU and also contrary to its duty to respect Article 4(2) TEU’.³⁶⁷ As the Corte constituzionale has emphasized:

There would be no respect if the requirements of unity demand the cancellation of the very core of values on which the Member State is founded. […] Otherwise, the European Treaties would seek, in a contradictory fashion, to undermine the very constitutional foundation of which they were born by the wishes of the Member States. […] It is therefore reasonable to expect that […] the European court will [leave] to the national authorities the ultimate assessment concerning compliance with the supreme principles of the national order.³⁶⁸

The interpretation of national identity under Article 4(2) TEU therefore does not provide an authoritative description of the boundaries of constitutional identity for the purpose of this study. The CJEU has no pure legal authority to interpret or loosen the bounds of constitutional identity by extra-constitutional interpretation.³⁶⁹ Whatever authority the CJEU has under Articles 19 and 4(2) TEU, it can only ever be derived authority from constitutional organs which themselves are subject to constitutional identity constraints on their conferring power.³⁷⁰

1.2.2.2 Normative Evaluation of Constitutional Identity Review

Pure constitutional claims aside, EU and national law also field competing normative descriptions of what values constitute constitutional identity, and subsequently what weight should be ascribed to them when they are in conflict with EU law. Normative disputes most frequently arise where the EU discovers itself to possess principles mirroring constitutional identity principles in national law, but these are interpreted differently, with a different rank, standard or content to the equivalent norm in national law. In particular, EU iterations of such norms can be subserviated to competing objectives of EU law – typically, the ‘effectiveness’ and ‘uniformity’ that justifies the supremacy of otherwise mundane acts of EU law. Member State constitutional identities cannot.

In that respect, although heterogeneous in specificity and entrenchment, Member State constitutional identities show a ‘remarkable convergence’ on two core normative principles:³⁷¹

• Constitutional Democracy, sometimes derived from popular sovereignty and sometimes from parliamentary or national sovereignty, is the basic principle of all Member State constitutions. The primary condition is that state law-making institutions remain accountable by election to the people in the manner specified in the constitution. Under all constitutional identity jurisdictions in this book, no state institution may validate an exercise of public power that is not democratically legitimated in the manner specified in the constitution. All, including the most basic among them, preclude a disposition of the Kompetenz-Kompetenz. The most developed, such as the German ‘eternity’ clause, entrench a specific formula for democracy: they require, in essence, that x powers can only be exercised by y institutions according to z formula, and these components themselves are not amendable.

• The Rule of Law requires that constitutional organs comply with substantive limits on state power inscribed in the constitution. This means that legislative and executive organs cannot transfer the power to act free from the constitution to the Union, because they are not empowered to act free from the constitution themselves.³⁷²

That these two principles can be essentially encapsulated as the definition of constitutional democracy is perhaps not surprising. And yet, the tension that arises whenever the ECJ interprets ‘national identity’ under Article 4(2) as having a different normative content or weight than national law seems to be a continuous source of surprise for Europe’s jurists. Indeed, some judges and scholars have poured scorn on the notion that the shape of the Union’s competences is constrained by the shape of national constitutional identities.³⁷³ The ECJ itself is under the duty to ‘respect’ national identities but has, by many accounts, made a hash of it.³⁷⁴ The ECJ has often refused to weigh constitutional identity considerations, even when flagged by AG Opinions,³⁷⁵ or the Member States themselves,³⁷⁶ and has sometimes dismissed or flatly ignored assertions from governments – and even constitutional courts – that some principle or other is part of the national identity.³⁷⁷ Despite several AG Opinions, Article 4(2) TEU was not cited in a single ECJ decision from its introduction in 1992 until after the rejection of the Constitutional Treaty in 2008.³⁷⁸ In the entire history of EU integration, just once has the ECJ found that a conflict between a fundamental constitutional right and an EU law compatible with the Charter could be decided in favour of the former.³⁷⁹ As Judge Pescatore has written, the teleology of CJEU interpretation is integration:

[T]he interpretation of Community Law depends not on the idea of maintaining an equilibrium which has been reached but on the vision of a European unity which is to be built.³⁸⁰

Accordingly, the EU courts are seen to have ‘laboured in the field of doctrine to extend the Community’s competences’,³⁸¹ to have ‘stretch[ed] their competences to the outermost limits [to] bring home the reality of European integration’,³⁸² and to evince a school of thought that ‘no opportunity should be missed of moving the Community caravan forward, if necessary by night marches’.³⁸³ Criticisms of a ‘dialogue among the deaf’ and a ‘lack of respect for the constitutional traditions of the Member States’ have been levelled against the ECJ where integral constitutional principles have been placed faithfully before it.³⁸⁴

Under Article 4(2) TEU, there is no recognition of inalienable constitutional reserves of sovereignty outside the legal order which can be invoked against the expansion of EU law.³⁸⁵ In all cases, ‘identity’ claims will be assimilated as ‘legitimate aims’ pursuant to a recognized EU derogation (and then subserviated to EU legislation under the proportionality test);³⁸⁶ or they will be assimilated as indistinguishable from EU norms – such as the protection of language or other fundamental values of the Union (and then interpreted in conformity with the EU law iteration).³⁸⁷ So, for example, in Melloni, it was accepted that the right to a fair trial under the Spanish Constitution could constitute national identity, but it was denied that it could be given a stricter interpretation than under the EU Charter.³⁸⁸ In Michaniki, AG Maduro explained:

[N]ational constitutional rules can be taken into consideration to the extent that they fall within the discretion available to the Member States […] within the limits fixed by the principle and the [instrument of EU legislation] itself.³⁸⁹

In short, constitutional identity is limited by the objectives of EU law, not the other way around.

The case for accepting this privileging of EU law over constitutional identity is normative: Member States must privilege the ‘uniformity and effectiveness’ of EU law over ‘constitutional identity’ claims, else the EU legal order will break down.³⁹⁰ The danger is what Kumm refers to as the ‘Cassandra scenario’ – constitutional identity review would cast the EU into inter-statal anarchy, threatening over sixty-eight years of peace and cooperation.³⁹¹ In Gauweiler v. Bundestag, AG Cruz-Villalón opined:

[I]t seems to me an all impossible task to preserve this Union, as we know it today, if it is to be made subject to an absolute reservation, ill-defined and virtually at the discretion of each of the Member States, which takes the form of a category described as ‘constitutional identity’ […] Such a ‘reservation of identity’, independently formed by the competent – often judicial – bodies of the Member States would very probably leave the EU legal order in a subordinate position.³⁹²

With respect, however, it is difficult to see why this is so, and virtually no constitutional court has accepted this normative justification over constitutional identity. For two reasons.

First, as Judge Kõve of the Riigikohus so puts it, ‘absolute’ supremacy would appear to ‘overestimate the theory’.³⁹³ Participation in this Union as we know it today simply does not entail ‘supranational “access” to the Member States’ legal orders’ outside its competences – particularly when no such authorization is even possible under many constitutions.³⁹⁴ That sort of ‘in for a penny, in for a pound’ argument has been dismissed as disingenuous and undemocratic.³⁹⁵ As the Trybunał Konstytucyjny has pointed out, ‘it is impossible in a democratic state ruled by law to create presumed competences’.³⁹⁶ The BVerfG agrees: ‘integration into a free community neither requires submission removed from constitutional limitation and control nor the forgoing one’s own identity’.³⁹⁷ The Corte constituzionale explains:

[W]hilst the aim of the [Corte constituzionale] is to preserve the constitutional identity of the Republic of Italy, it does not however compromise the requirements of uniform application of EU law [because it] does not result from an alternative interpretation of EU law, but exclusively from the fact, which in itself falls outside the substantive scope of EU law, that the Italian legal system […] subjects [criminal offences] to the principle of legality.³⁹⁸

Second, constitutional courts anyway doubt the normative superiority of a principle of legal ordering where the only inviolable principle is the effectiveness of executive-made law.³⁹⁹ In Fragd, for example, the Corte constituzionale stated that compared to the infringement of a fundamental principle, ‘concerns of uniform application of Community law and legal certainty did not have any overriding force’.⁴⁰⁰ Likewise, in Grogan (Ireland) the Irish Supreme Court stated:

Where an injunction is sought to protect a constitutional right, the only matter which could properly be capable of being weighed in a balance against it would be another constitutional right […] there can be no question of a possible or putative right which might exist in European law as a corollary to a right to travel so as to avail of services, counterbalancing [that right] as a matter of convenience.⁴⁰¹

Simply put, Member State constitutional courts do not weigh EU laws and constitutional identity norms in accordance with the normative weight the ECJ ascribes to them.⁴⁰² The architects of EU fiscal federalism can place no stock in the claim that the ‘effectiveness and uniformity’ of EU law will impel constitutional courts to accept intrusions on constitutional identity in order to accommodate some ideal institutional model. They are not authorized to decide that EU objectives should persist while national constitutional guarantees should perish.⁴⁰³

1.2.2.3 Positivist Evaluation of Constitutional Identity Review

Finally, the merits for accepting the absolute supremacy of EU law as a positivist statement of the law governing the boundaries of ‘constitutional identity’ are dubious. This was demonstrated in recent cases such as Weiss (Germany),⁴⁰⁴ Slovak Pensions XVII (Czech Republic),⁴⁰⁵ R v. Oberlandesgericht (Germany),⁴⁰⁶ Ajos (Denmark),⁴⁰⁷ Taricco II (Italy),⁴⁰⁸ and HS2 (UK),⁴⁰⁹ where national courts in fact invalidated or disapplied ECJ rulings, and these decisions were in fact taken as an authoritative statement of law by the legal system.

Indeed nearly every constitutional court – even the most communautaire among them – has invalidated or interpreted EU law in conformity with the boundaries of national constitutional identities, rather than the other way around. If this has averted newsworthy open conflicts most of the time, it has nonetheless led to a diffusive realm of ‘parallel’ interpretations where EU law is nonetheless invalidated or warped against the shape of constitutional identities. This can be seen in, inter alia, Constitutional Treaty (France),⁴¹⁰ EM Eritrea (UK),⁴¹¹ Gauweiler (Germany),⁴¹² HS2(UK),⁴¹³ Anti-terror Database (Germany),⁴¹⁴ Pham v. SSHD (UK),⁴¹⁵ AAA v. Strix (Sweden),⁴¹⁶ Constantinou (Cyprus),⁴¹⁷ the Portuguese financial conditionality cases,⁴¹⁸ Grogan (Ireland),⁴¹⁹ ESM (Estonia),⁴²⁰ Sugar Quotas III (Czech Republic),⁴²¹ EAW (Poland),⁴²² Riga Land Use (Latvia),⁴²³ Data Retention (Romania),⁴²⁴ Telecommunications Market Act (Finland),⁴²⁵ Money Laundering (Belgium),⁴²⁶ DI.KATSA (Greece),⁴²⁷ Auxiliary Activities in the Public Sector (Croatia),⁴²⁸ Agricultural Surplus Stocks (Hungary),⁴²⁹ and Taricco II (Italy),⁴³⁰ where courts exercised a sort of ‘reverse-Simmenthal’ supremacy or studiously ignored conflicting interpretations of EU law entirely.

Article 4(2) TEU may therefore be said to constitute a ‘material’ (merely persuasive) competence to blunt an EU measure before it protrudes over the boundaries of the EU legal order into constitutional identities, but it is clear it does not have ‘formal’ authority – Member States do not accept the supremacy of the ECJ’s assessment over their own.⁴³¹ To the contrary, where the CJEU has asserted itself over constitutional identity adjudication, the jurisdiction has proven so constitutionally fraught that its very use is prejudicial the integrity of the European legal order (it should not be forgotten that it was precisely that phenomenon in Internationale Handelsgesellschaft which provoked the birth of ‘constitutional identity’ jurisprudence in the first place).⁴³²

Most recently, the rather transparent attempt to absorb constitutional identities into the EU legal order under Article 4(2) TEU is credited with provoking the emergence of new ‘constitutional identity’ jurisprudence in Belgium,⁴³³ and the recentralization of overlapping EU Charter/constitutional claims in the Constitutional Courts of Italy and Austria.⁴³⁴ In Austria, the VfGH recently asserted that the interpretation of EU rights ‘must heed the constitutional traditions of the Member States and therefore the distinct characteristics of the rule of law in the Member States’.⁴³⁵ In Italy, the Corte constituzionale recently reasserted its jurisdiction to ‘ensure that the rights [under the EU Charter] are interpreted in a way consistent with constitutional traditions’.⁴³⁶

Prior to that, the straight assertion of supremacy over constitutional identity in Åkerberg Fransson (CJEU)⁴³⁷ and Melloni (CJEU)⁴³⁸ provoked a broader rebellion to EU supremacy in Melloni (Spain),⁴³⁹ Taricco II (Italy),⁴⁴⁰ R v. Oberlandesgericht (Germany),⁴⁴¹ Anti-terror Database (Germany)⁴⁴² and HS2 (UK),⁴⁴³ where constitutional courts attacked the ECJ’s reasoning and reasserted their own supreme constitutional principles over EU law.⁴⁴⁴

In R v. Oberlandesgericht (Germany), the BVerfG explicitly rebuffed Melloni under its Solange I (Germany) jurisdiction, overturning a decision of a lower court even though ‘the [lower] Court’s decision is determined by Union law’ and the ECJ had ‘specifically ruled’ that execution of a warrant could not be conditional on compliance with constitutional law.⁴⁴⁵

Melloni also received a drubbing under the controlimiti doctrine in Taricco II (Italy), forcing the CJEU into a volte-face after the Corte constituzionale held, again contra Melloni, that ‘the Italian Constitution construes the principle of legality in criminal matters more broadly than European law’,⁴⁴⁶ and an ECJ interpretation of the TFEU contrary to that standard ‘therefore may not be permitted, even in light of the primacy of EU law’.⁴⁴⁷

Similarly, in Anti-terror Database (Germany), the BVerfG refused to submit a preliminary reference as obliged by the ECJ’s Åkerberg Fransson decision (seen by some as an extension of EU competence) and appeared to state that Åkerberg Fransson was itself ultra vires and inapplicable in Germany:

The ECJ’s decision in the case Åkerberg Fransson […] must not be read in a way that would view it as an apparent ultra vires act […] in a way that questioned the identity of the Basic Law’s constitutional order.⁴⁴⁸

The UK Supreme Court followed suit in HS2 (UK), where it refused to submit a reference on the compatibility of a ‘Hybrid Bill’ process with EU law, and held that ‘a decision of the [ECJ] should not be read by a national court in a way that places in question the identity of the national constitutional order’.⁴⁴⁹

More recently, in Ajos (Denmark), a unanimous decision of the Højesteret refused to disapply a provision of domestic legislation as directed by the ECJ because to apply the principle of age discrimination ‘as interpreted by the EU Court of Justice’ would be contra legem.⁴⁵⁰

In 2012, the straight application of supremacy to Czechoslovakian dissolution arrangements in Landtová v. Česká správa socialního zabezpečení provoked a constitutional identity ruling by the Ústavní Soud so vociferous it bears full repetition here:

[The Ústavní Soud] expected that […] the ECJ would familiarize itself with the […] constitutional identity of the Czech Republic, which it draws from the common constitutional tradition with the Slovak Republic [idem est] a completely idiosyncratic and historically-created situation that has no parallel in Europe. […]

The failure to distinguish legal relationships arising from the dissolution of a state with a uniform social security system from legal relationships arising from the free movement of persons in the European Communities […] is a failure to respect European history; it is comparing matters that are not comparable. For this reason it is not possible to apply European law […] it is not possible to do otherwise than to find […] that an act ultra vires has occurred.⁴⁵¹

As the European Law Journal editors wryly point out, EU primacy vis-à-vis the national pouvoir(s) constituant(s) grants the ECJ ‘a power that perhaps can only exist as long as it is not made use of’.⁴⁵²

A power that can ‘perhaps exist as long as it is not made use of’ cannot offer an authoritative statement of law for the purposes of this study. Constitutional courts have stated (and demonstrated) that legal architectures will be invalidated if they exceed EU competence or intrude on constitutional identities, and this study must take them at their word. Member State Kompetenz-Kompetenz and constitutional identity jurisprudence provides a valid constitutional, normative and positivist descriptions of the limits of the EU legal order for the purposes of this study on fiscal federalism.

1.3 The Constitutional Boundaries of European Fiscal Federalism

The conclusion that Member State Kompetenz-Kompetenz and constitutional identity jurisprudence provides a valid description of the constitutional boundaries of the EU legal order means the architects of fiscal federalism cannot look solely to EU law, as interpreted by the CJEU, as the ultimate constraint on European fiscal federalism. Member State constitutional courts impose constraints not only on the current boundaries of EU law lex lata, but also on potential expansions of EU law and revisions of the EU Treaties de lege ferenda.⁴⁵³ The question of whether a specific fiscal federalism model might ‘work’ in the EU must heed these fundamental constitutional limits of European integration in national constitutional law. That being so, the remainder of this chapter will attempt to specify the precise substantive boundaries which will impinge upon the selection of fiscal federalism systems in the EU.

1.3.1 Fiscal Sovereignty

The first boundary is Member State fiscal sovereignty. This principle is implicitly but plainly impressed upon the allocation of competences in economic policy (Articles 2(3) and 5(1) TFEU) and the substantive provisions governing public finance (Articles 121–126 TFEU). Under those articles, the EU has no competence in economic and fiscal policy.⁴⁵⁴ The Union competence under these articles is ‘mere coordination’,⁴⁵⁵ limited to providing ‘a framework to coordinate these policies to a certain degree’.⁴⁵⁶ This is not a mere reflection of good administration under the principle of subsidiarity (though it undoubtedly coheres with that principle).⁴⁵⁷ As the BVerfG so puts it, fundamental decisions on public finance and expenditure are ‘a fundamental part of the ability of a constitutional state to democratically shape itself’, ‘the core of parliamentary rights in democracy’ and ‘an essential manifestation of constitutional democracy’.⁴⁵⁸

This marks an immutable boundary of the EU legal order. Not only has economic and fiscal policy not been conferred on the Union, but, according to the BVerfG, it cannot ever be so conferred without abrogating the national constitutional identity and violating the ‘eternity clause’ (Article 79(3)) of the 1949 German Basic Law.⁴⁵⁹ In Lisbon (Germany), it held:

A transfer of the right of the Bundestag to adopt the budget and control its implementation by the government [would] violate the principle of democracy […] in its essential content.⁴⁶⁰

Numerous other courts have drawn similar boundaries around national fiscal sovereignty. In Lisbon (Poland) the Trybunał Konstytucyjny held that the conduct of ‘independent financial, budget and fiscal policies’ is one of the ‘attributes of sovereignty’ comprising Poland’s constitutional identity.⁴⁶¹ In Crotty (Ireland) the Irish Supreme Court stated that the freedom to form economic policy ‘is just as much a mark of sovereignty’ as the freedom to legislate itself, such that the desire to ‘qualify, curtail or inhibit the existing sovereign power […] is not within the power of the Government itself’.⁴⁶² In Collins (Ireland) the High Court held that ‘Budgetary allocation is a fundamental responsibility which [the] Constitution cast upon the Daíl […] This constitutional responsibility may under no circumstances be abrogated, whether by statute, parliamentary practice or otherwise.’⁴⁶³ In TSCG (France) the Conseil Constitutionnel held that Articles 120–126 TFEU did not ‘infringe the essential conditions for the exercise of national sovereignty’ because they did ‘not result in the transfer of any powers over economic or fiscal policy’.⁴⁶⁴ In TSCG (Belgium), the Cour constitutionnelle held that public finance measures belong to the ‘democratically elected legislative assembly, solely competent for this purpose’ and ‘[i]t is therefore up to the respective parliaments to exercise this budgetary competence’.⁴⁶⁵ The Spanish Tribunal Constitucional holds that budgetary autonomy is the essence of ‘the ability to self-government, expressed especially in the possibility of developing [a region’s] own policies or matters within their range of competence’.⁴⁶⁶ In Sweden, parliamentary fiscal competences are listed among the Basic Principles in Chapter 1 of the Instrument of Government excluded from conferral under Chapter 10§6.⁴⁶⁷ In Lithuania the Konstitucinis Teismas holds that decisions concerning state loans and liabilities ‘may be adopted by the Seimas only […] an institution [which] may neither transfer nor waive these powers. Such powers may neither be changed nor limited by law’.⁴⁶⁸ In EFSF (Slovenia), the Ustavno Sodišče held that ‘the fundamental power of the National Assembly […] to decide on state revenue and expenditure’ fell under the ‘principle of a state governed by the rule of law and the principle of the legality of the operation of the state administration’ (Slovene constitutional identity) and so could not be delegated to another institution, including the executive.⁴⁶⁹ In Latvia the Satversmes tiesa holds that, ‘the law on the state budget is an important function of the Saeima, which it fulfils as an institution directly responsible to the people of Latvia’⁴⁷⁰ and ‘solely the legislator can take decisions concerning the state budget’ under the basic principles of the democratic state.⁴⁷¹ In Croatia, the Ustavni sud holds that ‘the exclusive authorities of the Government and the Croatian Parliament concerning issues relevant for the State Budget’ are part of the ‘constitutional identity’ beyond the reach of amendment by referendum.⁴⁷²

In a string of 2011 rulings on the constitutionality of the EFSF/ESM legal frameworks before the Irish Supreme Court,⁴⁷³ the German BVerfG,⁴⁷⁴ the Austrian VfGH,⁴⁷⁵ the Finnish Perustuslakivaliokunnan,⁴⁷⁶ the Polish Trybunał Konstytucyjny,⁴⁷⁷ the Estonian Riigikohus⁴⁷⁸ and the Slovenian Ustavno Sodišče,⁴⁷⁹ the legality of the EFSF or ESM were predicated on the conclusion that financial commitments were capped to the extent of the parliamentary authorization, so the agreements did not entail an open-ended transfer of fiscal sovereignty. In ESM (Estonia) the Riigikohus explained:

The sovereignty of the people gives rise to the sovereignty of the state and thereby all state institutions obtain their legitimation from the people. […] One element of the state’s sovereignty is its financial sovereignty, which contains taking decisions on budgetary matters and on the assumption of financial obligations for the state.⁴⁸⁰

In all countries which have had occasion to pronounce on the matter in the context of EU integration, parliamentary control over fiscal policy is what separates a (constitutional) exercise of sovereignty from an (unconstitutional) abrogation of constitutional identity.⁴⁸¹

1.3.1.1 Three Tests for Fiscal Sovereignty

This book extracts three tests for evaluating whether a proposed legal arrangement coheres with the limits of Member State fiscal sovereignty under European constitutional identity case law:

• [1.3.1.2] A restriction on budgetary sovereignty must not ‘fetter the budget legislature to such an extent that the principle of democracy is violated’, i.e., ‘with the effect that it or a future Parliament can no longer exercise the right to decide the budget on its own’;⁴⁸²

• [1.3.1.3] A delegation of budgetary decision-making must not compromise the principle that ‘the [national] Parliament remains the place in which autonomous decisions on revenue and expenditure are made’;⁴⁸³ and

• [1.3.1.4] A finite financial disposition must not be of structural significance to the Parliament’s right to decide on the budget such that it causes an irreversible prejudice to future majority decisions and cannot be reversed by an equivalent action by the Parliament in the future. The test applied is that ‘the democratic process remains open and that legal re-evaluations may occur on the basis of other majority decisions and that an irreversible legal prejudice to future generations is avoided’.⁴⁸⁴

Although the burgeoning Member State case law on these principles appears remarkably convergent thus far, it must be said that these tests are quarried, first and foremost, from the leading German jurisprudence, and it is that jurisprudence which this section will expound upon to explain these tests. This is so for two reasons.

First, much of the legal architecture at issue in this book derives directly from German constitutional constraints. Price stability (Article 127(1) TFEU), the prohibition on monetary financing (Article 123 TFEU), the ‘no bailout’ rule (Article 125 TFEU) and the fiscal governance rules (Articles 121–126 TFEU) are ‘parallel provisions’ to the German Basic Law, and ‘permanent constitutional requirements of German participation in the monetary union’.⁴⁸⁵

Second, the ‘eternity clause’ that grounds the German ‘constitutional identity’ jurisdiction is unusually strong and well-defined compared to other ‘identity’ provisions in Europe. It is the high-water mark of constitutional identity in Europe – and it is unamendable. While novel instruments proposed for EU fiscal federalism may trespass on constitutional identity in any number of countries, they will most likely cross the limits of Article 79(3) BL first. Article 79(3) states:

Amendments of this Constitution affecting the division of the Federation into Länder, their participation in principle in the legislative process, or the basic principles laid down in Articles 1 [Human Dignity] and 20 [Democratic and Social Federal State] shall be inadmissible.

This provision is a permanent feature of German – and European – constitutional heritage. It is, according to the BVerfG, an indelible consequence of history – ‘a reaction to the historical experience of a creeping or abrupt erosion of the free substance of a democratic fundamental order’.⁴⁸⁶ It permanently shields the highest constitutional principles of the German state – human dignity (Article 1 BL)⁴⁸⁷ and the basic principles of the democratic social and federal State (Article 20 BL)⁴⁸⁸ – from constitutional change.

Fiscal sovereignty falls primarily within the protection of the basic principles of the democratic and social federal state under Article 20. Article 20 states, in part:

(1) The Federal Republic of Germany is a democratic and social federal state.

(2) All state authority emanates from the people. It is exercised by the people through elections and voting and by specific organs of the legislature, the executive power and the judiciary.⁴⁸⁹

The principles of popular sovereignty and constitutional democracy in Article 20(2) secure the constitutional link between the act of voting in elections and the exercise of state power. As stated by the BVerfG: ‘Article 20(2) sentence 2 guarantees in conjunction with art.79(3) that the exercise of state duties and the exercise of state powers can be traced back to the people of the state and are accounted for vis-à-vis the people.’⁴⁹⁰

This is in turn given substance by the right to vote in Article 38.⁴⁹¹ Article 38 states, in part:

(1) The deputies to the German House of Representatives [Bundestag] are elected in general, direct, free, equal and secret elections. They are representatives of the whole people not bound by orders and instructions, and subject only to their conscience.

(2) Anyone who has attained the age of eighteen years is entitled to vote; anyone who has attained majority is eligible for election.

The right to elect the Bundestag under Article 38(1) is a right to elect a parliament that remains accountable to the people which elect it.⁴⁹² This precludes legal commitments entered into by treaty ‘if the result of this is that the people’s democratic self-government is permanently restricted in such a way that central political decisions can no longer be made independently’.⁴⁹³ In ESM (Germany), the Court held:

A necessary condition for the safeguarding of political latitude in the sense of the core of identity of the constitution (art.20(1)-(2), art.79(3) BL) is that the budget legislature makes its decisions on revenue and expenditure free of other-directedness on the part of the bodies and of other Member States of the European Union and remains permanently ‘the master of its decisions’.⁴⁹⁴

Any break in the ‘chain of legitimation’ between the right to vote under Article 38(2) and the exercise of state power under Article 20 will prima facie constitute an infringement of German constitutional identity under Article 79(3). If voters are no longer able to exercise the right to vote under Article 38(2) BL; if the right to vote is to be exercised by a method of voting other than the formula described in Article 38(1); if votes are no longer connected to the autonomous Bundestag in Article 38(1); or if the Bundestag no longer possesses the substance of the power to rule through conferral or ‘other-directedness’ (Article 20(2) BL) – then the chain of legitimation will be broken.⁴⁹⁵ What is guaranteed under the German Constitution is not just ‘democracy’ in an openly defined or purely formal sense.⁴⁹⁶ It is ‘self-determination in the exercise of public power’ through a specific democratic formula.⁴⁹⁷ It is the substance of the power to rule:

Article 38 [BL] protects the citizens with a right to elect the Bundestag from a loss of substance of their power to rule, which is fundamental to the structure of a constitutional state, by far-reaching or even comprehensive transfers of duties and powers of the Bundestag, above all to supranational institutions.⁴⁹⁸

Under Article 79(3) BL, the basic principles and constituent structures of the democratic social and federal state are inviolable.⁴⁹⁹ They may not be weighed against any other legal interests (including the mandate of peace and integration and the constitutional principle of the openness towards EU law);⁵⁰⁰ they cannot be narrowed or disposed of by constitutional amendment;⁵⁰¹ and they cannot be weighed against the ‘constructive force of the mechanism of integration’.⁵⁰² They cannot be transcended in the name of public good under a Schmittian state of exception,⁵⁰³ and so cannot be interpreted in the light of effet utile or ultima ratio justifications seen to underlie recent EU crisis measures – no matter how meritorious.⁵⁰⁴ As the BVerfG has stated, Article 79(3) does not require ‘cases of imminent totalitarian seizure of power’ for it to be exceeded.⁵⁰⁵ Indeed, it is precisely that argument which Article 79(3) is meant to guard against.⁵⁰⁶ In Lisbon (Germany), the BVerfG held:

The principle of democracy may not be weighed against other legal interests; it is inviolable. The constituent power of the Germans which gave itself the Basic Law wanted to set an insurmountable boundary to any future political development. Amendments to the Basic Law affecting the principles laid down in art.1 and art.20 of the Basic Law shall be inadmissible.

(art.79.3 of the Basic Law)⁵⁰⁷

How, then, is EU legislation to be squared with that formula? EU Parliamentary elections are not taken in the general, direct, free and equal manner prescribed by Article 38(1) BL; it is not the German people in Article 38(2) BL which exercise state power through the Bundestag in Article 38(1); and the European Parliament, the Council and the Commission are not the legislature and executive in Article 20(2) BL.⁵⁰⁸

The answer is that, within the context of the EU, constitutional identity is safeguarded by Article 23(1) BL. It states:

To realize a unified Europe, Germany participates in the development of the European Union which is bound to democratic, rule of law, social, and federal principles as well as the principle of subsidiarity and provides a protection of fundamental rights essentially equivalent to that of this Constitution. The federation can, for this purpose and with the consent of the Senate [Bundesrat], delegate sovereign powers. Article 79(1) & (3) is applicable for the foundation of the European Union as well as for changes in its contractual bases and comparable regulations by which the content of this Constitution is changed or amended or by which such changes or amendments are authorized.⁵⁰⁹

This constitutional safeguard clause creates an ‘exception’ for democratic opinion-forming in ways different to that envisioned under Article 38 BL, but this only ‘applies as far as the limit of the inviolable constitutional identity’ of which Article 20 and its machinery (Article 38) are a part.⁵¹⁰ In short, powers conferred on the union can be conferred up to the hilt of Article 79(3), but no further. In Lisbon (Germany), the BVerfG explained:

The empowerment to embark on European integration permits a different shaping of political opinion-forming than the one determined by the Basic law for the Constitutional order. This applies as far as the limit of the inviolable constitutional identity (art.79.3). […] The minimum standard protected by art.79.3 of the Basic Law must not fail to be achieved even by Germany’s integration into supranational structures.⁵¹¹

In Lisbon (Germany), the BVerfG enumerated a list of inalienable, essential powers so ‘particularly sensitive for the ability of a constitutional state to democratically shape itself’ that they comprise the substance of self-government and fall under the umbrella of the eternity clause.⁵¹² These included fiscal competences, criminal law, monopoly of force, social living conditions, and decisions of cultural importance, such as family, education and religion.⁵¹³ Fiscal policy was among the most important of those powers. The BVerfG held:

Particularly sensitive for the ability of a constitutional state to democratically shape itself are […] fundamental fiscal decisions on public revenue and public expenditure. […] A transfer of the right of the Bundestag to adopt the budget and control its implementation by the government [would] violate the principle of democracy and the right to elect the German Bundestag in its essential content if the determination of the type and amount of the levies imposed on the citizen were supranationalised to a considerable extent. The German Bundestag must decide, in an accountable manner vis-à-vis the people, on the total amount of the burdens placed on citizens. The same applies correspondingly to essential state expenditure. In this area, the responsibility concerning social policy in particular is subject to the democratic decision-making process, which citizens want to influence through free and equal elections. […] What is decisive, however, is that the overall responsibility, with sufficient political discretion regarding revenue and expenditure, can still rest with the German Bundestag.⁵¹⁴

From this and the case law which follows, this book extracts three ways by which fiscal sovereignty may be denuded in violation of Articles 38, 20 and 79(3) BL.

1.3.1.2 Unlawful Restrictions on Fiscal Sovereignty

The first way in which the principle of democracy might be denuded is through formal restrictions on parliamentary budgetary powers, ‘with the effect that it or a future Bundestag can no longer exercise the right to decide the budget on its own’.⁵¹⁵ As representatives of the people under Article 38(1), not bound by any orders or instructions, the Bundestag ‘must retain control of fundamental budgetary decisions even in a system of intergovernmental administration’.⁵¹⁶ If the German Bundestag were to find itself in the role of ‘mere subsequent enforcement’, it could ‘no longer exercise its overall budgetary responsibility’.⁵¹⁷ In Euro Rescue Package, the BVerfG stated:

[F]undamental decisions on public revenue and public expenditure are part of the core of parliamentary rights in democracy. Article 38.1 excludes the possibility of depleting the legitimation of state authority and the influence on the exercise of that authority provided by the election by fettering the budget legislature to such an extent that the principle of democracy is violated.⁵¹⁸

It should be emphasized that it is not, from the outset, undemocratic for the budget-setting executive to be fettered by a particular fiscal policy. In ESM II, the BVerfG accepted that a commitment to a particular fiscal policy may be made through agreeing corresponding obligations under international law.⁵¹⁹ The test for evaluating whether a fetter on budgetary autonomy amounts to an unconstitutional deprivation of sovereignty is whether control over that policy is relinquished, such that the fetter is not reversible by an equivalent act of the Bundestag in the future.⁵²⁰ The test applied is that ‘the democratic process remains open and that legal re-evaluations may occur on the basis of other majority decisions and that an irreversible legal prejudice to future generations is avoided’.⁵²¹

1.3.1.3 Unlawful Conferral of Fiscal Sovereignty

The second way the substance of the power to rule might be depleted is through delegation or conferral of the powers of the parliament itself.⁵²² The budgetary powers still exercised by the parliament must not be depleted to such a degree that the right to make legal re-evaluations of budgetary policy under Articles 38 and 20 BL is rendered meaningless.⁵²³ The test in that regard is the same: A violation of the principle of democracy will occur ‘if the German Bundestag relinquishes is parliamentary budget responsibility with the effect that it or a future Bundestag can no longer exercise the right to decide on the budget on its own responsibility’.⁵²⁴ In Euro Rescue Package, the BVerfG held:

The relevant factor for adherence to the principles of democracy is whether the German Bundestag remains the place in which autonomous decisions on revenue and expenditure are made, even with regard to international and European commitments.⁵²⁵

First and most obviously, this means the parliament cannot confer its competence in budgetary policy. A violation of the principle of democracy would occur if ‘the type and amount of the levies imposed on the citizen were supranationalised to a considerable extent and thus the Bundestag would be deprived of its right of disposal’.⁵²⁶

Second, Articles 38 and 20 BL cannot simply be got-around by signing over the common finances of the citizenry by blank cheque. The Bundestag may not transfer its budgetary responsibility through ‘imprecise authorisations’ or mechanisms with ‘incalculable burdens’ that are tantamount to accepting liability for decisions by free will of other states.⁵²⁷ The BVerfG has explicitly precluded the ‘transfer union’ or ‘liability community’ and instruments of loss-sharing in which budgetary dispositions are no longer determined by the autonomous exercise of the free will of the Bundestag in the manner required by Article 38 BL.⁵²⁸ In Euro Rescue Package, the Court held:

The Bundestag may not transfer its budgetary responsibility to other actors by means of imprecise budgetary authorisations. In particular it may not, even by statute, deliver itself up to any mechanisms with financial effect which […] may result in incalculable burdens with budget relevance without prior mandatory consent, whether these are expenses or losses of revenue. […] The Bundestag must specifically approve every large-scale measure […] involving public expenditure on the international or European level.⁵²⁹

1.3.1.4 Unlawful Impairments of Fiscal Sovereignty

Finally, even a finite disposition must not be so large that the Bundestag is no longer able to conduct economic policy on its own responsibility.⁵³⁰ The right to vote under Article 38 would be as equally meaningless if the Bundestag elected to give over the entire endowment of the citizenry, in one lump sum, as it would be if it signed up to open-ended authorizations.

However, on this limb, the BVerfG exercises a high degree of curial deference where finite dispositions are concerned. The test applied to finite dispositions is a ‘manifest overstepping of ultimate limits’⁵³¹ – that is, whether the amount of the disposition is ‘of structural significance for parliament’s right to decide on the budget, for example by giving guarantees the honouring of which may endanger budget autonomy’.⁵³²

In monetary terms, the Court has refrained from putting a number on this ‘ultimate limit’, but it seems nothing short of over half the federal budget will do. In Euro Rescue Package (Germany), the pledging of a sum ‘far greater than the largest federal budget item’ and ‘substantially exceeding half of the federal budget’ did not deprive the Bundestag of its autonomy.⁵³³ In ESM I, budget commitments of €190,024,800,000 (approximately 50% of all central government expenditure) did not exceed the legislature’s margin of appreciation, so long as it did not constitute an open-ended commitment and did not deprive the parliament of the ability to shape the economic and social life of the state.⁵³⁴

There is a ceiling to this, however. In Weiss (Germany), the BVerfG held that risk-sharing through the PSPP, ‘which amounts to more than EUR 2 trillion […] would affect the limits set by the overall budgetary responsibility of the German Bundestag […] and be incompatible with Art. 79(3)’.⁵³⁵

1.3.1.5 Permissible Limitations on Fiscal Sovereignty

From this case law, this book extracts the above-noted three tests (listed at the start of Section 1.3.1.1) for determining whether fiscal sovereignty is infringed under the leading German constitutional identity jurisprudence. This does not mean that the contours of other countries’ jurisdictions are not lurking just behind.⁵³⁶ However, given that the German tests are likely to remain the leading tests in this area, it is useful to highlight some room for manoeuvre through permissible limitations on fiscal sovereignty under these tests. Contrary to how Article 79(3) BL is sometimes perceived, ‘constitutional identity’ does not mean that all the core constitutional powers are absolutely and forever entombed at national level, with no capacity for delegation. There are three limits on the jurisdiction.

First, the words ‘particularly sensitive’ in Lisbon (Germany) indicate that not all ‘state-founding elements’ are included in the list of competences listed in that decision, and not all intrusions to that list will violate Article 79(3).⁵³⁷ It is only if the competence is both particularly sensitive and the formula for democratic legitimation specified in the constitution is structurally compromised that constitutional identity is infringed.⁵³⁸ For example, the expansion of QMV under the Lisbon Treaty did not infringe constitutional identity because the scope of conferral was controlled under Article 23 BL, and the essential powers under the umbrella of Article 20 were still exercised in accordance with Article 38 BL.⁵³⁹

Second, the enumeration of constitutional identity competences in Lisbon (Germany) does not mean that those core competences can never be delegated; it means that they cannot be conferred or delegated in a manner which breaks the chain of legitimation under the German constitution. There is a difference. For example, automatic budgetary liability under the ‘capital calls’ provisions of the ESM Treaty did not violate Article 38 BL, because the voting formula gave Germany an effective veto over each new disposition to the ESM. Similarly, monetary policy is lawfully conferred on the ECB because the conditions which apply to the ECB under Article 127 TFEU are the same as those that apply to the Bundesbank under Article 88 BL, so no usurpation of fiscal competence could occur.⁵⁴⁰ The essential staple is that delegation is permitted, as long as this does not change the substance of the guarantee itself.⁵⁴¹

Third, not all encroachments on ‘state founding’ powers will constitute a violation of democracy in its essential content.⁵⁴² For fiscal policy, this will only occur where a fiscal policy decision is not reversible by an equivalent action by the Bundestag and the degree of the infringement is of structural significance to Parliament’s right to decide on the budget.⁵⁴³ So, for example, we know from Weiss (Germany) that €2 trillion is too much, but in ESM (Germany), the Court applied a test of proportionality and a margin of discretion to huge sums – approximately 50% of all central government expenditure – without this constituting a complete failure of budgetary autonomy.

1.3.2 Price Stability and Fiscal Discipline

The second constitutional boundary of European fiscal federalism pursued in this book is comprised of the fundamental guiding principles of price stability, sound public finances and a sustainable balance of payments binding on the mandate for EMU under Article 119(3) TFEU. These are the principles of the ‘Stabilitätsgemeinschaft’ or ‘Stability Community,’ which limit the mandate for monetary union and define the decentralized model of fiscal federalism inscribed in the Treaties. Article 119(3) TFEU reads:

These activities [economic and monetary policy] of the Member States and the Union shall entail compliance with the following guiding principles: stable prices, sound public finances and monetary conditions and a sustainable balance of payments.

The first principle is price stability. Price stability is the first constitutional principle of EMU and the sole objective of EU monetary policy competence.⁵⁴⁴

The second and third guiding principles, sound public finances and sustainable balance of payments, are principles of fiscal and economic policy – competences of the Member States.

Sound public finances means that Member States must run a sound fiscal policy that avoids excessive public debts or sovereign defaults with adverse spillovers on monetary policy.

Sustainable balance of payments means they must run a sound economic policy so that the external deficit of the country as a whole does not become unsustainable, impoverishing the country and leading to the same result.

Hereafter, this book generally refers to these two principles together under the single term ‘fiscal discipline’.⁵⁴⁵

These principles, price stability and fiscal discipline, inform the entire legal architecture of fiscal federalism under Articles 119–127 TFEU. The design of this architecture is discussed in Chapter 2, but it is sufficient to remark here that the principles of the Stabilitätsgemeinschaft are a constitutional stipulation of the EU’s conferred competence in monetary policy and economic coordination. As stated in Brunner (Germany):

Article [119 TFEU] sets up the guiding principles for member-States’ activities the maintenance of price stability, sound public finances and monetary conditions, and a sustainable balance of payments. […] This conception of the currency union as a community based on stability is the basis and subject-matter of the German Act of Accession. If the monetary union should not be able to develop on a continuing basis […] within the meaning of the agreed mandate for stabilization, it would be abandoning the Treaty conception.⁵⁴⁶

The fundamental principles of the Stabilitätsgemeinschaft have been linked by the BVerfG to the independence of the ECB,⁵⁴⁷ price stability,⁵⁴⁸ the prohibition on monetary financing,⁵⁴⁹ the ‘no bailout’ clause⁵⁵⁰ and the Stability and Growth Pact.⁵⁵¹ In particular, the BVerfG has warned that the principles of Stabilitätsgemeinschaft would be violated – in turn violating Articles 20 and 79(3) of Germany’s constitutional identity – if the Union should become a ‘liability community’ through the ‘direct or indirect communitarisation of state debts’.⁵⁵²

This section will explain how these principles reflect deeper constitutional boundaries underlying the EU legal order as a whole.

1.3.2.1 Price Stability

Under Articles 3(1)(c), 119(2) and 127 TFEU, and Articles 2–3 and 17 to 24 of the Statute of the ESCB, the ECB’s monetary policy competence and all of the ECB’s instruments are bound to the primary objective of price stability (defined as 2% inflation by the ECB Governing Council). Subject to that objective, it may also ‘support’ economic policies which contribute to the aims of the Union, but it can pursue none of its own.⁵⁵³

This, too, is a restriction carved directly from the German Basic Law.⁵⁵⁴ Article 88 BL states:

The Federation establishes a note-issuing currency bank as the Bundesbank. Its tasks and powers can, in the context of the European Union, be transferred to the European Central Bank which is independent and primarily bound by the purpose of securing stability of prices.⁵⁵⁵

Article 88 permits conferral of monetary competence on the ECB only in so far as it remains independent and bound to price stability. Unlike the Bank of Canada,⁵⁵⁶ the Bank of England⁵⁵⁷ or the United States Federal Reserve,⁵⁵⁸ for example, the ECB can have no mandate for financial stability. Not because the EU legislator would not allow it, but because the German Basic Law does not allow the German legislator to confer it.

Since Brunner v. EU Treaty (Germany), the primacy of price stability has been central to the constitutionality of Germany’s ongoing participation in the EMU under Article 79(3)BL.⁵⁵⁹ The BVerfG has held, for instance, that ‘The Union Treaty governs the monetary union as a community which is permanently obliged to maintain stability and, in particular, to guarantee the stability of the value of the currency.’⁵⁶⁰ A development contrary to that mandate would violate the conditions subject to which monetary policy was conferred, mandating ‘withdrawal from the Community in the event of the community based on stability failing to materialise’.⁵⁶¹

Article 88 BL is not, in and of itself, part of the German constitutional identity shielded by the eternity clause in Article 79(3) BL. Article 88 could be amended and it would pose no further constraint on conferral. An ordinary breach of that provision will first fall to BVerfG’s ultra vires review jurisdiction, under which the BVerfG will afford a margin of appreciation to an ultra vires act unless it is ‘structurally significant’ to the division of competences.⁵⁶²

However, Article 88 does shield other constitutional provisions which are linked to Article 79(3) BL. These are, specifically, the right to property under Article 14 (protected by Article 1 BL), which guards against the expropriation of value from money-holders through inflation; and the basic principles of the democratic state under Article 20 BL, which protects the constituent power against unauthorized or open-ended financial dispositions.⁵⁶³

The reason Article 88 shields these principles is that, unlike federal banks in Canada, the United States or Switzerland, the funding structure of the ECB has the potential to circumvent parliamentary control of budgetary policy. This is because the ECB is financed by all EMU Member States in accordance with the ESCB capital key. This is unlike the Bank of Canada, the United States Federal Reserve and the Swiss National Bank, which are not financed by the contributions of their provinces, states or cantons. When the United States Federal Reserve conducts bond purchase operations, for example, it purchases the bonds of a separate federal treasury, independently of state treasuries. The bonds are not guaranteed by any state governments, and so ‘The Fed is not bailing out a cash-strapped country [and] distributing risks among the taxpayers with an excellent credit rating.’⁵⁶⁴ In the United States, ‘the printing presses cannot be used to provide particular states or regions with credit at below-market interest rates’,⁵⁶⁵ and purchases of public sector securities ‘do not lead to redistributional effects among the individual states of the US’.⁵⁶⁶

In the EU, by contrast, deliberately targeting the bonds of, say, Greece would use taxpayer contributions from all countries to assume risks incurred by one country and, as the Bundesbank states: ‘Monetary policymakers have no authorisation to redistribute such risks or burdens among the taxpayers of various euro-area countries.’⁵⁶⁷ Because the Bundestag backstops the Bundesbank, an expenditure campaign by the ECB for an economic objective – like bond market stability or staving off state defaults – would commit parliamentary funds to an economic policy without a parliamentary vote. For this reason, a violation of Articles 123 or 127 TFEU will not only be ultra vires Article 88 BL, but may constitute a structurally significant infringement of constitutional identity.⁵⁶⁸

1.3.2.2 Fiscal Discipline: Sound Budgetary Policies and a Sustainable Balance of Payments

In the field of economic policy, the principles of ‘fiscal discipline’ – sound budgetary policy and a sustainable balance of payments – manifest in the legal architecture under Articles 119–126 TFEU. That architecture is examined in Chapter 2, however it suffices to state here that the model entrenches independent financial liability and the budgetary autonomy of national parliaments. For this reason, these provisions are also constitutional stipulations of Germany’s participation in EMU. As stated in Weiss Reference (Germany):

The current European integration agenda is based on an understanding of the monetary union as a community of stability; for [Germany], this is an essential prerequisite for its membership in the monetary union. Most notably, this safeguards the German Bundestag’s overall responsibility for the budget.⁵⁶⁹

It should be emphasized here, too, that while fiscal discipline and Articles 121–126 TFEU safeguard the German constitutional identity, ‘not every single manifestation of the stability community is guaranteed by [Article 20 BL] in conjunction with art.79(3)’.⁵⁷⁰ Violations are first and foremost a matter of ultra vires review, not constitutional identity, unless it also violates one of the tests set out in Section 1.3.1 of this book.

In practice, however, it may make no difference how many lines are crossed since a violation of the Stabilitätsgemeinschaft that results in automatic financial liability or deprives parliamentary control over fiscal policy will also lead to a violation of Articles 38, 20 and 79(3) BL, and the consequences of both ultra vires and identity review are invalidity.⁵⁷¹ So, for example, as a matter of economics, a failure to achieve budgetary discipline implies monetary financing or debt mutualization, and this offends the right to property (Article 14 BL) and the right to vote (Article 38 BL), which are part of the constitutional identity in conjunction with Article 1 BL (Human Dignity) and Article 20 BL (Basic Principles), and are not amendable under Article 79(3) BL. Hence, even if no individual act of fiscal indiscipline will vitiate the Stabilitätsgemeinschaft, the overall system of fiscal federalism chosen for the EMU must be based on fiscal discipline and individual financial responsibility if it is to ultimately remain within its constitutional boundaries. However, unless the three tests set out in Section 1.3.1 are also met, the test applied here is different: It is whether the Union violated the ‘community based on stability (Stabilitätsgemeinschaft) [that] is the basis and subject-matter of the German Act of Accession […] within the meaning of the agreed mandate for stabilisation’.⁵⁷²

1.4 Conclusions: Permanent Constraints on European Fiscal Federalism

The constitutional boundaries extracted in this chapter are real, they are permanent, and they exert real positive force on the boundaries of EU law. Constitutional courts have stated (and demonstrated) that nascent machineries of fiscal federalism will be invalidated if they trespass on constitutional fiscal sovereignty or exceed the boundaries of conferral, and this study must take them at their word. This conclusion derives from three cumulative analyses.

[1.1] First, the EU is a ‘federation of states’, possessed of a top-down federal hierarchy with a legal supremacy greater than any individual expression of Member State sovereignty on one hand, yet on the other hand derived from the confederate authority of national orders which sanction its reach. However, the reality that concerns this book is that, whether one adopts a Kelsenian, normative or positivist approach, national constitutions (as interpreted by national constitutional courts) remain the reference point for validity of law in Member State legal systems.

[1.2] In the EU, national constitutional orders profess to impose two limits on the EU’s conferred powers: First, that they have the jurisdiction to assert, through Treaty ratification and ultra vires review, what powers they have and have not conferred on the Union – the so-called Kompetenz-Kompetenz. Second, that their own ‘constitutional identity’ principles determine the absolute limits of Union law. These assertions pose a valid constitutional, normative and positivist description of the limits of the EU legal order.

[1.3] Under these jurisdictions, two substantive constitutional boundaries will bear upon any model of European fiscal federalism. [1.3.1] The first is Member State fiscal sovereignty. Not only have parliamentary competences in economic and fiscal policy not been conferred on the Union, but, according to the BVerfG, they cannot ever be so conferred without abrogating the ‘Basic Principles’ of the ‘Democratic State’ (Article 20) and violating the ‘eternity clause’ (Article 79(3)) of the 1949 German Basic Law. Numerous other constitutional courts have drawn similar boundaries around fiscal sovereignty.⁵⁷³ The tests applied by this book in that regard are:

• [1.3.1.2] No unlawful restrictions of fiscal sovereignty: A restriction on budgetary sovereignty must not ‘fetter the budget legislature to such an extent that the principle of democracy is violated’, that is, ‘with the effect that it or a future Parliament can no longer exercise the right to decide the budget on its own’;⁵⁷⁴

• [1.3.1.3] No unlawful conferral of fiscal sovereignty: A delegation or conferral of financial competences must not compromise the principle that ‘the [national] Parliament remains the place in which autonomous decisions on revenue and expenditure are made’;⁵⁷⁵ and

• [1.3.1.5] No structural impairments of fiscal sovereignty: even a finite financial disposition must not structurally impair the parliament’s right to decide on the budget and shape the economic and social life of the state in the future.⁵⁷⁶

[1.3.2] The second constitutional boundary is comprised of the fundamental guiding principles of price stability and fiscal discipline (sound budgetary policy and sustainable balance of payments) impressed upon the architecture in Articles 119–127 TFEU. Articles 119–127 TFEU are not in themselves part of Member State ‘constitutional identity’; however, the architecture of the Stabilitätsgemeinschaft indirectly shields basic principles of the democratic state (Article 20 BL) and human dignity (Article 1 BL), which are part of the constitutional identity shielded by the German ‘eternity clause’ and are not amendable, lex lata or de lege ferenda.

Having identified these principles underlying the boundaries of the EU legal order in economic and monetary policy, Chapter 2 will seek to examine how they inhere in the legal architecture inscribed in the EU Treaties as a matter of EU law.