1  The institutions of the European Union (EU or ‘Union’) evolved in a historical process. The shape of the original institutions at the moment of foundation of the European Communities—the European Coal and Steel Community (ECSC), the European Economic Community (EEC), and the European Atomic Energy Community (Euratom)—in the 1950s was due to the then prevailing political framework. This created path dependencies which were partly followed, partly complemented by new such dependencies in a series of institutional arrangements (Merger Treaty, Act concerning the election of the representatives of the Assembly by direct universal suffrage), and Treaty reforms (Single European Act, Treaty of Maastricht, Treaty of Amsterdam, Treaty of Nice, Treaty of Lisbon).

2  The Treaty of Lisbon emphasized that the institutional framework of the EU has to fulfil the requirements of a democratically sound regime (see Democracy in the EU). According to Article 10(1) of the Treaty on European Union (‘TEU’), ‘[t]he functioning of the Union shall be founded on representative democracy’. Article 10(2) TEU specifies the basic content of such regime: ‘Citizens are directly represented at Union level in the European Parliament’. At EU level, democratic legitimacy emanates from the Union citizens, represented in the European Parliament (Art. 14(2), first subsection, first sentence TEU; see European Parliament: Composition and Election – EU Law). Shortcomings in the democratic composition of the European Parliament are therefore bound to lead to a democratic deficit. The requirements of representative democracy are further specified in Article 10(4) TEU providing the basis for the establishment of European Political Parties. Article 11 TEU, particularly its section 4 on the Citizens’ Initiative, intends to strengthen democracy in Europe adding an element of direct democracy to representative democracy.

3  Article 10(2) TEU continues in its second subsection: ‘Member States are represented in the European Council by their Heads of State or Government and in the Council [of the EU] by their governments, themselves democratically accountable either to their national Parliaments, or to their citizens’. The representation at EU level within the European Parliament is thus complemented by the requirement of democratic representation at Member States’ level. This dual democratic representation is advantageous for the EU. Democratic rule in the Member States is—recent challenges in some Member States notwithstanding—settled and firm. In integrating the representative democracy grown in the Member States, the EU strengthens its own legitimacy. In this regard, ‘National Parliaments contribute actively to the good functioning of the Union (Art. 12 TEU).

4  This second, Member State-related pillar of democratic rule in the EU hints at the fact that the EU is not only a representative democracy, but also exhibits strong elements of a federation. Theorists struggle about a reasonable terminology for such federation sui generis, which is not a federal State like the United States (US), Canada, Australia, Switzerland, or Germany. Sometimes, the aspect of associating sovereign States is highlighted (Staatenverbund), at other instances, the emphasis is on associating constitutions (Verfassungsverbund: multilevel constitutionalism), and there is a trend to merge these theoretical approaches. In constitutional practice, it is the Member States that confer competences to the EU to attain common objectives (Art. 1(1) TEU); competences are thus distributed between the Union and the Member States (Arts 4 and 5 TEU). Therefore, the representation of the Member States in the European Council (Art. 15 TEU) and the Council of the EU (Art. 16 TEU) is crucial and follows the institutional logic of federalism. Article 10(2), second subsection TEU clarifies that the Member States’ representation in these institutions has to be democratic. This provision is open to democratic accountability of governments either towards national parliaments or towards the electorate directly—as in presidential democracies, eg France.

5  The EU strives for accomplishing democratic rule in its federal structure by means of the so-called Community—a reference to the predecessor of the EU, the European Communities (EC). This label is used for the cooperative framework of the EU’s three core institutions, the European Parliament, the European Council, the Council of the EU (‘Council’), and the European Commission (‘Commission’), all mentioned on top of the enumeration in Article 13(1) TEU. The general idea is that legislation, executive, and budgetary powers are not vested in an intergovernmental arrangement but in a complex triangle between parliamentary representation, democratic representation from the Member States, and the Union interest in an abstract sense. Together, they shall accomplish the normative obligations from Article 10(1) and (2) TEU and the values and aims described in Articles 2 and 3 TEU.

6  The European Parliament is the central institution to provide for democratic representation of the Union citizens within the EU. The extension of the idea of parliamentary democracy to the supranational sphere of the entire EU is at the same time a tremendous achievement from a historical perspective and enormous progress towards attaining democratic rule beyond the State on a global scale. The European Parliament had a very modest start as the Assembly of the ECSC, composed of MPs from the Member States’ parliaments and meeting only once a year to discuss the High Authority’s annual report. Only the lack of space led to its transferral from Luxembourg to Strasbourg, which is still its (formal) seat. Over time, it gained competence after competence, first in budgetary and later in legislative matters. The idea of a direct election, first realized in 1979, was a decisive step to advance parliamentarism in the EU.

7  Today, a series of core functions is vested in the Parliament (see European Commission: Tasks, Functions and Powers). It co-legislates together with the Council and these two institutions also share the budgetary powers of the EU (Art. 14(1) first sentence TEU). These two functions alone are a strong realization of the democratic principle within the federative structure of the EU as laid down in Article 10(2) TEU. What is more, the European Parliament guarantees the democratic legitimacy of the Commission (see European Commission, Composition and Investiture). The President of the Commission is elected by the European Parliament (Arts 14(1) third sentence and 17(7) second subsection, second sentence TEU) which goes beyond the practice under former treaties only asking the Parliament for its assent to the selection of the Member States. Such assent is still necessary for the entire Commission (Art. 17(7) third subsection, first sentence TEU), which is also accountable towards the European Parliament (Art. 17(8) TEU with Art. 234 Treaty on the Functioning of the European Union (‘TFEU’); cf Art. 226 TFEU for Committees of Enquiry and Art. 233 TFEU for the debate on the Commission’s annual report). Sometimes, the European Parliament has been described as the main winner of the various Treaty reforms leading to the Treaty of Lisbon. Indeed, the legal parliamentarization of the EU is impressive. Today, representative democracy in the EU means parliamentary democracy. The involvement of the national parliaments (Art. 12 TEU) does not weaken but strengthens this finding (see National Parliaments in the EU).

8  Whether or not we summarize them under the political heading ‘democratic deficit’, there continue to be shortcomings in the European parliamentary democracy which create a need for reform. If such reform fails, the whole construction of democratic rule in the EU is imperilled. Nearly all of the shortcomings lie in the composition of the European Parliament and the law on how it is elected (see European Parliament, Composition and Election).

9  According to Article 14(3) TEU, MEPs are elected by direct universal suffrage in a free and secret ballot. The equality of votes is missing, and this not without reason. The seats within the European Parliament (maximum 750, plus President, Art. 14(2), second sentence TEU) are distributed according to a scheme which is ‘degressively proportional, with a minimum threshold of six members per Member State’. Also, ‘[n]o Member State shall be allocated more than ninety-six seats’ (Art. 14(2) second and third sentence TEU). The Treaty provision already petrifies inequality of voters: if the traditional slogan ‘one man, one vote’ was applied, the by population largest Member State would have to be 16 times bigger than the smallest. The contrary is obvious: the largest Member State, Germany, has 83 million inhabitants, the smallest are Malta, 0.43 million, Luxemburg, 0.55 million, and Cyprus, 0.86 million. Representation is overtly disproportionate following the 96:6 ratio. The relevant European Council Decision 2018/937/EU based on Article 14(2) second subsection TEU hardly alleviates this disproportionality—which is circumscribed in the Treaty by ‘degressively proportional’.

10  The ratio behind degressive proportionality is of course the protection of the representative interests of smaller Member States, which is legitimate per se. Taking a comparative view, one can discern similar disproportionalities for the same reason in other federal entities that are even vested with statehood, as in the US (Art. I(2) third section, third sentence US Constitution requires a minimum of one representative per state in the House of Representatives). Furthermore, there is some compensation of that inequality by the vote in the co-legislating Council which takes into account the size of the population of the various Member States (Art. 16(4) TEU), while not being based on the population size alone. Finally, the nationality of MEPs does not play a central role in voting in the European Parliament. However, the deficit in representation is significant despite these possible justifications. It should be addressed. Reform options include the creation of transnational voting lists or transnational constituencies. It can only be deplored that courageous proposals in this respect were finally rejected in preparation of the 2019 elections that had opened—albeit for the reason of Brexit—the door to the experimental application of such reform proposals.

11  Thus far, the institutions did not make proper use of their legislative powers to pass ‘provisions necessary for the election of [the Parliament’s] Members by direct universal suffrage in accordance with a uniform procedure in all Member States or in accordance with principles common to all Member States’ (Art. 223(1) TFEU). This is to be criticized all the more as Art. 223 TFEU is one of the few provisions where the European Parliament has a right of initiative. True, even in the US elections at national level are still partly governed by the law of the states. However, the divergences of the different voting rules in the EU are so substantial that the non-transparency of democratic representation threatens the Parliament’s legitimacy. The harmonization has also failed hitherto to establish a common minimum threshold for accessing the Parliament. The proposal to establish this threshold at 3% failed in 2018, with the German Federal Constitutional Court not playing a helpful role in this story. The over-representation of small, often extremist, and sometimes nonsense parties does not advance parliamentarism in the EU.

12  The concept of political parties at European level is still not viable. Although there are families of national parties at the European level, the electorate still votes for national and not European parties. This brings the whole election process closer to national than to European politics. This is one of the reasons for the low voter turnout, although it rose from 2014 to 2019.

13  The Council mirrors the supranational democratic position of the European Parliament on the federal side. It provides for the democratic backup of the Member States’ representation within the EU. As such, the Council is the co-legislator with the European Parliament and shares budgetary powers with it (Art. 16(1) first sentence TEU). Its core task beyond legislation and budget is described in Article 16(1) second sentence TEU as: ‘[i]t shall carry out policy-making and coordinating functions as laid down in the Treaties’. In this regard, the Council is the link between the Member States and the Union as a whole, and the details of that function are laid down in the various Treaty provisions enabling the Council to act (see Council of the European Union: Tasks, Functions and Powers).

14  The voting method of the Council has been settled since the Treaty of Lisbon, following a long and difficult development (see Council of the European Union: Composition). From the beginning, every Member States has been represented in the Council by one member at ministerial level, which leaves the question of sound representation to the law of the Member States (Art. 16(2) TEU: ‘who may commit the government of the Member State in question and cast its vote’). Based on this composition, the Council is one institution in different configurations depending on the deliberated policy field. Two of these configurations are foreseen in the Treaty, the General Affairs Council and the Foreign Affairs Council (Art. 16(6) subsection 2s and 3 TEU), the former being in charge of coordinating the whole Council, preparing the European Council meetings and ensuring their follow-ups. Other configurations are adopted by the European Council according to Article 16(6) subsection 1 TEU and Article 236(a) TFEU. Council Decision 2009/878/EU now lists ten configurations, some of which are very visible in public such as the Ecofin Council or the Agriculture and Fisheries Council. In matters of the Eurozone, the right to vote in the Ecofin Council may be restricted to the Member States concerned (Art. 136(1) and (2) TFEU). The Presidency rotates among Member States (Art. 16(9) TEU, Art. 236(b) TFEU) except for the Foreign Affairs Council which is presided by the High Representative of the Union for Foreign Affairs and Security Policy (Art. 18(3) TEU).

15  Deliberation and voting in the Council carry the risk of opaque and unequal intergovernmental-diplomatic negotiations—a method from the past not adhering to the requirements of representative democracy. In fact, this picture is part of the memory about the EU: marathon meetings, bilateral talks in the ‘confessional box’, clocks stopped to match deadlines, and the like. In law, the Lisbon Treaty, prepared by preceding Treaty reforms, brought about major progress against those shortcomings. Voting is generally by qualified majority, and qualified majority is clearly defined as at least 55% of the members of the Council comprising at least 15 of them and representing Member States comprising at least 65% of the population of the Union. A blocking minority—in a Union of 27—includes at least four Council members; if this is not achieved the qualified majority shall be deemed attained (Art. 16(3) and (4) TEU). Furthermore, legislative deliberation in the Council is public (Art. 16(8) TEU).

16  Nonetheless, the negotiating rationale is still dominant. Member States introduce their interests into the process, and the outcome is often a compromise. Consensus is the rule, a formal vote is the exception. Renovating such compromise, ie changing the law, is more difficult in a negotiation-based setting. Furthermore, in the recent past, the Council developed a growing administrative body. Council decisions have always been prepared by expert groups from the Member States whose deliberations were streamlined by COREPER (Comité des représentants permanents) before entering high politics. In the expert groups, the negotiation of national bureaucracies is organized. The Council administration provides for a more structured and solid shape of this decision-making. The size of the Council’s administrative background is even reflected architecturally: in Brussels, the Justus LipsiusBuilding—hosting the Council—equals the Berlaymont where the Commission is lodged.

17  The very supranational, sui generis-European player in this context is the European Commission. In law, it has always been the neutral institutional safeguard and promotor of the general interest of the EU (Art. 17(1) TFEU). Its historical roots are not democratic, not in a sense of anti-democracy or worse, but of bureaucracy. Jean Monnet, one of the founding fathers of European integration and first President of the High Authority of the ECSC, had been head of the French planning authority and other similar institutions before the creation of the High Authority which later mutated into the Commission. We know that the independent regulatory commissions that had blossomed during the New Deal in the US had exemplary functions for the creation of the High Authority: independent bureaucracies distant from day-to-day politics and established to institutionally guarantee a particular output, often in the field of economics. It was only by intervention of the smaller Member States—Benelux at the time—that the Council was created; the original idea had been to shape the ECSC around the High Authority only. Throughout the history of European integration, the Commission was the engine of the ‘ever closer Union’—Article 1(2) TEU in the version of the Lisbon Treaty, formula used since the Treaty of Maastricht, similar formula used since the Treaty of Rome—initiating important steps, the most famous being certainly the internal market, laid down in a Commission white paper in 1985 and accomplished in 1992.

18  It is part of the Community method that the Commission preserved this bureaucratic tradition of expertise and orientation towards the general interest of the EU. The quality of performance of the Commission’s services is still impressive; the competitive system of recruiting all over Europe guarantees a significant level of expertise. In law, this particular position of the Commission is defined by its ‘monopoly of initiative’ (Art. 17 (1) first sentence TEU, Arts 292 and 293 TFEU), meaning that the Commission has the sole right to propose legislation in the vast majority of areas. The Commission is the guardian of the Treaties, implements the budget, holds the main administrative functions, and is in charge of the external representation of the Union except in the Common Foreign and Security Policy (CFSP) (Art. 17(1) TEU). Its procedure is governed by the collegiate principle: it decides by a majority of its members (Art. 250(1) TFEU). Its departments—mentioned in Article 249(1) TFEU and specified, albeit not in detail, in Article 21 ff. Rules of Procedure of the European Commission—comprise about 33,000 officials (see European Commission, Internal Organisation).

19  However, it would be more than wrong to say that the Commission had no democratic legitimacy (see European Commission: Composition and Election). To the contrary, it is part of the developed Community method that the President is elected by the European Parliament by a majority of its component members (Art. 17(7) first subparagraph, second sentence TEU) and that the whole college of the Commission has to be approved by the European Parliament as well (Art. 17(7) third subparagraph TEU). The influence of the European Parliament upon the investiture of the Commission, in particular the appointment of the President, and the President’s role within the Commission has been constantly growing throughout the various Treaty reforms. The high-water-mark of that development, though, was the political idea of nominating top politicians for each political family running for the European election in 2014 as candidates for the Commission Presidency, Spitzenkandidaten—a rare non-military word from Germany to conquer political language in Europe. With some resistance—from Hungary and the United Kingdom, above all—this idea was politically successful in 2014 and Jean-Claude Juncker was proposed by the European Council and elected by the Parliament. Although some were arguing the contrary, there was never a legal obligation from the phrase ‘[t]aking into account the elections to the European Parliament’ (Art. 17(7) first subparagraph, first sentence TEU) to propose one of the Spitzenkandidaten. The political logic of such proposal, however, has been obvious ever since. But as there was no clear majority for one of those candidates in 2019, the absence of legal obligation became visible.

20  Another effect of the Spitzenkandidaten-process and the strengthening of the Commission President was the idea to give the Commission a stronger political profile. Juncker explicitly declared its Commission a ‘political Commission’. This led to a streamlining of the college’s structure, with groups of Commissioners around the vice-presidents, devoted to the subject-matters of the Commission’s priorities. Politically, this advanced the position of the Commission in a considerable way. Other instances of the idea of a political Commission were less helpful, such as the abandonment of the deficit procedure (Art. 126 TFEU) against France ‘puisque c’est la France’. Apparently, this imperilled the credibility of the neutral safeguard of the Treaties. Ursula von der Leyen’s actual Commission is not led by a former Spitzenkandidat, as Manfred Weber had been rejected for lack of political experience and as the European People's Party’s (EPP) ‘victory’ was not as clear—and thus not as accepted by the Socialists & Democrats (S&D)—as in 2014. Von der Leyen continued organizing and shaping in a similar way as Juncker; yet her rhetoric has not been similarly prominent so far, which is probably due to the challenges of the pandemic c that hit the new Commission soon after coming into office.

21  The view on the institutional framework of the Union would be incomplete with at least a glimpse on its Court of Justice, although its composition, procedure, and detailed functions must be treated elsewhere. In the institutional context, the Court’s task ‘to ensure that in the interpretation and application of the Treaties the law is observed’ (Art. 19(1) second sentence TEU) must be highlighted. All institutions of the EU must abide by the legal provisions laid down in the Treaties. Likewise, all Member States are bound by the Treaties. Hence, control of compliance with the Treaties is not left to the political sphere or the Member States as such but to a common court of law composed of independent judges and a distinct institutional framework to assure the maintenance of the rule of law in the EU.

22  The clear picture of the Community method—including its modifications and the shortcomings of its realizations—is still quantitatively dominating the work of the EU. However, the progress enshrined in that method is challenged by some institutional developments that do not just come as political irritations but as clear institutional change, mainly constitutionalized in the Treaties.

23  The European Council has been accompanying the integration process as part of European political cooperation since the early 1970s. It was first recognized at Treaty level in the Single European Act and institutionalized as a separate institution and not just as a particular configuration of the European Council. The European Council is composed ‘of the Heads of State or Government of the Member States, together with its President and the President of the Commission’ (Art. 15(2) first sentence TEU), and its function is to ‘provide the Union with the necessary impetus for its development and [it] shall define the general political directions and priorities thereof’ (Art. 15(1) first sentence TEU). Although the European Council is not vested with legislative functions (Art. 15(1) second sentence TEU) it is somehow placed on top of the institutional pyramid and, in times of crisis, its impulses are sometimes very near to legislating (eg the European Council meeting as Eurozone summit according to Art. 136(1) and (2) TFEU). The European Council may have a norm-creating function in the simplified amendment procedure under Article 48(6) TEU (hitherto applied once, Art. 136(3) TFEU). Furthermore, within the European Council, international treaties between the Member States, which accompany EU law, can be negotiated. Although all this is not legislation in a formal sense, functionally, it resembles legislation and thus is in potential conflict with the prohibition to legislate. The European Council decides by consensus, thus emphasizing intergovernmentalism, diplomatic procedures, and sometimes opaqueness.

24  The sheer existence of the European Council shows the continuing intergovernmental basis of the EU. It was further strengthened by the Treaty of Lisbon when the office of the President of the European Council was created, who is elected by a qualified majority for a maximum of two terms of two and a half years each (Art. 15(5) first sentence TEU). The European Parliament is not involved in the election of the President of the European Council; its (democratic) legitimacy is thus only drawn from the Member States. The relationship between the President of the European Council and the President of the Commission—and also the Council Presidency—is not settled by constitutional law but left to the political sphere, which leads to instances of Presidents acting as a team—sometimes together with the President of the European Parliament and/or the President of the European Central Bank. Historically, the establishment of a President for the European Council is reported to go back to September 2001 when the then holder of the rotating presidency of the Council, the Belgian Guy Verhofstadt, was unable to reach the White House by phone. Some heads of state advanced the idea in the European Convention and the following intergovernmental conference in the mid-2000s. The institutional reality shows, however, that the President’s main task is less the external representation as a ‘President of Europe’ but rather the balancing of Member States’ interests. This is best illustrated by the choice of two Belgians, Herman van Rompuy and Charles Michel, who had respective experience from their very complicated federal background but were less well-known on the international stage. Also, Donald Tusk held this moderating role. At any rate, this political reality underlines the European Council’s intergovernmental basis, which cannot conceal that this institution has the potential to weaken the Community method. Politically, this was verbalized when German Chancellor Angela Merkel spoke of the ‘Union method’ to replace the Community method, in particular in the field of economic policy in 2010. Finally, it should be noted that a double-hat—the President of the European Council and of the Commission being one and the same person—is not explicitly excluded under the Treaties, although its validity is subject of controversy.

25  Foreign affairs and defence have never been part of the Community method, though narrow exceptions linked to the intergovernmental EU—sanctions—and limited fields of the internal market—procurement in military matters—exist. The creation of a coordinating function, somebody Henry Kissinger could call by phone, was bound to follow the logic of the intergovernmental common foreign and security policy. It is therefore not astonishing that the High Representative of the Union for Foreign Affairs and Security Policy is, upon consent by the President of the Commission, appointed by the European Council (Art. 18(1) TEU) and takes part in its work (Art. 15(2) second sentence TEU). The problem of multiple loyalties notwithstanding, the integration of the High Representative into the institutional framework of the EU by making them at the same time Vice-President of the Commission (Art. 18(4) first sentence TEU) and permanent chairperson of the Foreign Affairs Council (Art. 18(3) TEU), supported by the European External Action Service (EEAS), was certainly a helpful move.

26  The enumeration in Article 13 TEU is intended to streamline the institutional framework of the EU which had not been systematic before. Institutional plurality is characteristic for European integration from its inception, starting with three Communities, initially all having separate institutions—ie High Authority plus the EEC and the Euratom-Commission. The—somewhat opaque—merger of these institutions through a special Merger Treaty created a simplified institutional framework. Unfortunately, clarity by simplification ended or was at least significantly reduced by the co-existence of the intergovernmental EU and the supranational EC since the Treaty of Maastricht. The Treaty of Lisbon and in particular Article 13 TEU put an end to that—albeit not entirely. Indeed, if the institutions of the Union were limited to the enumeration of that Article, the arrangement of internal organization would be rather clear. In addition to the European Parliament, the European Council, the Council, and the Commission, Article 13(1) TEU mentions three other institutions: the European Court of Justice, the Court of Auditors, and the European Central Bank.

27  Article 13(4) TEU designates two auxiliary bodies having advisory functions: the European Economic and Social Committee is a ‘dinosaur’ from the founding period of the ECSC and EEC, copied from the French Conseil économique et social, bringing together stakeholders from various economic fields—trade unions, employers, etc—and being without any visible effect in EU legislation. A similar statement can be made about the Committee of the Regions. The Treaty of Maastricht established this body giving voice to the regional and local level and thus compensating for their loss of power. The composition of the Committee, however, is too diverse to be institutionally powerful: mayors, members of regional governments, and diets with deviating parochial interests and of different political capabilities. Both Committees prove that institutional extensions should be created with care—abolishing superfluous bodies seems to be significantly more difficult than creating them.

28  Maybe, this should be seen as a warning with respect to a mega-trend in European institutional development: agencification. Beginning in the 1970s with bodies of minor importance and also limited powers, agencies began to blossom quantitatively following the Commission’s white paper on European governance. Two main categories have to be discerned.

29  Executive Agencies—eg the Education, Audiovisual and Culture Executive Agency (EACEA) or the Research Executive Agency—have their own budget responsibility but remain in the institutional sphere of the Commission. They were established to face the growing responsibility for very large financial support programmes. The concomitant highly technical and specialist tasks of management and financial control should be transferred to the executive agencies rather than outsourced to private consultancies and bodies which had occasionally led to mismanagement and even fraud in the past.

30  Regulatory Agencies—eg Agency for the Cooperation of Energy Regulators (ACER), European Banking Authority (EBA), or Frontex—follow a different concept: they are created in addition to the Commission services for the fulfilment of specific administrative tasks. Their boards—managing bodies with varying designations—are composed of Member States’ representatives, at times complemented by a Commission representative. As some Regulatory Agencies grew out of Member States’ networks to become separate bodies, this form of agencification is a phenomenon of European federalism. In this regard, EU Regulatory Agencies differ from US agencies—or independent regulatory commissions—whose institutional ratio has nothing to do with federalism but with the dichotomy between the President and the Congress. It is also a sign of EU federalism that the Regulatory Agencies are located all over Europe.

31  Initially, Regulatory Agencies had few competences beyond the preparation of delegated/implementing legislation for the Commission, the collection of data, or just factual and scientific advice for EU politics. This changed with the financial crisis around 2010 when the financial supervisory agencies (EBA, European Securities and Markets Authority (ESMA), and European Insurance and Occupational Pensions Authority (EIOPA)), were created. These agencies were vested with real powers, from case-by-case decision-making to the adoption of general rules. With this rise in powers the question of their limits became pertinent again. As a matter of principle, the Court of Justice of the European Union (CJEU) adheres to case-law from the 1950s, the Meroni judgments, which require a clear circumscription of powers delegated to agencies and prohibit the delegation of discretionary powers.

32  This case law has hitherto not been able to tame agencification. The challenge is to maintain institutional coherence in an EU with about 40 agencies spread all over the Union. The Treaty (in particular Article 298 TFEU) is still underdeveloped for this task, but Treaty change is out of reach. Legal oversight must be strengthened with a particular role for the Commission as the main administrative body. The requirement of independence, highlighted in Article 298 TFEU and laid down in the founding regulations of the agencies, is opposed to political oversight and accountability towards the European Parliament. Whatever the intensity of Member States’ influence, agencies remain part of the EU administration and must therefore be better integrated into its institutional framework.

33  A significant step in building the current institutional design of the EU was the creation of the European Central Bank (ECB) in the 1990s. Whereas the Commission is somehow weakened by the rise of intergovernmental elements such as the European Council Presidency, by the conclusion of international agreements between Member States instead of the application of the Community method as well as by institutional pluralization, the supranational power of the ECB is ever growing. This is due, on the one hand, to monetary and political necessities: a common currency needs a strong central bank, and the transfer of powers for the supervision of the financial sector lies within this monetary and political logic. On the other hand, there are clear legal provisions guaranteeing the ECB’s special position.

34  First and foremost, the ECB is vested with a particular form of independence to safeguard the fulfilment of its primary objective which is the maintenance of price stability (Art. 282(2) second sentence TFEU). Consequently, the ECB ‘shall be independent in the exercise of its powers and in the management of its finances. Union institutions, bodies, offices and agencies and the governments of the Member States shall respect that independence’ (Art. 282(4) third and fourth sentence TFEU). Such independence is not a novelty in European integration as the Commission is designed as an institution independent of the Member States as well, but it is unique in the world of central banks which are usually under a certain political influence. Even the German Bundesbank pre-Maastricht had no constitutional independence, although its statutory and political independence was significant and served as a role model for the ECB. The ECB’s independence is problematic in terms of democratic legitimacy but justified by its clear constitutional objective: price stability. The ECB shall support the general economic policies in the Union (Arts 119(2) and 127(1) TFEU). However, if the ECB would pursue its own aims of economic policy, for which it has no competence, it would not only run the risk of acting ultra vires, but also of being a stakeholder outside the requirements of democratic legitimacy laid down in Article 10(1) and (2) TEU. This is at the core of the debate about the ECB’s powers in financial supervision. Independence from Member States’ governments is crucial as their influence on eg the standards of supervision would significantly destabilize the European Banking Union. The ECB’s tasks concerning financial supervision are, however, somehow distant to its primary objective of price stability. Therefore, methods ensuring democratic accountability while preventing political manipulation should be sought.

35  Furthermore, the supranational character of the ECB is underlined by its institutional design. The ECB is part of the European System of Central Banks (ESCB) comprising also the national central banks of the EU. They, too, are required to be independent and thus shielded from political influence at Member States’ level. The Governing Council of the ECB, which is at the same time the Governing Council of the ESCB, is composed of the members of the Executive Board of the ECB and the governors of the central banks of the Eurozone Member States. To simplify the decision-making process within the Governing Council, voting rights are rotating among the governors of the Eurozone Member States. The Executive Board is composed of the President, the Vice-President, and four other members.

36  The particular supranational independence of the ECB provided for its enormous institutional strength in the financial and state debt crises since 2008. However, the integration of the ECB into the democratic institutional structure of the EU remains a significant challenge. The second, yet significantly less powerful, bank founded by the EU is the European Investment Bank whose task and mandate is ‘to contribute, by having recourse to the capital market and utilising its own resources, to the balanced and steady development of the internal market in the interest of the Union’ (Art. 309 TFEU). In particular, the Bank grants loans and gives guarantees to finance projects for less-developed regions, modernizing and converting undertakings and projects of common interest to several Member States.

37  The institutional arrangement of the EU is able to perform the ordinary functions of public power, although there is reluctance in using the usual terminology. This is particularly true for the legislative function. The non-ratified Treaty establishing a Constitution for Europe had contained provisions on ‘legislative acts’ defined as ‘European laws and framework laws’ in its Article I-34. After the failure of its ratification, Member States agreed to avoid all state-analogous terminology, thus also abandoning such terminology as ‘European law’ in the Treaty of Lisbon. It was, however, upheld in substance: remainders of the legislation-terminology are still visible in the Treaty text, although in a less comprehensible way. In reality the rule-making function of the EU is a legislative function. Therefore, ‘legal acts’, according to the relevant TFEU section, are legislative acts if they are meant to establish general rules for the whole EU.

38  The textual ambiguity of the Treaty does not only lie in shifting the provisions on legislation to the TFEU despite their overall importance. All ‘legal acts adopted by legislative procedure (Art. 289(3) TFEU) are legislative acts. They are complemented by delegated and implementing acts. In addition, Article 288 TFEU still uses the traditional wording—regulation; directive; decision; recommendation; opinion—to distinguish European acts. In theory, the possible combinations create a framework of nine binding forms of legal acts—the non-binding recommendations and opinions set aside. Enhanced-Cooperation, which may be used as ultima ratio to legislate for only parts of the EU, complicates this system even further.

39  Legislative acts are subdivided into two categories according to the legislative procedure leading to their adoption. In specific cases provided for by the Treaties, the special legislative procedure (Art. 289(2) TFEU) is applicable. In these cases, legal acts are created by the European Parliament with the participation of the Council (eg Art. 223(2) TFEU) or by the Council with the participation (ie consent or hearing) of the European Parliament (eg Art. 19(1) or 127(6) TFEU). Usually, however, the ordinary legislative procedure applies (Art. 289(1) TFEU). On a proposal from the Commission, the European Parliament and the Council jointly adopt a legislative act acting as co-legislators according to the institutional provisions. The designation of the co-legislative procedure as ordinary is the high-water-mark of the Parliamentary influence in EU legislation which took decades to be reached. But this legal picture is flawed by factual developments: legislation is usually negotiated at the first reading-stage between the three stakeholders, the European Parliament, Council, and Commission. In these so-called trilogues—the linguistic validity of the extrapolation of dialogue to trilogue being questioned more than often—representatives of the three institutions meet to prepare the formal acts of adoption: the relevant Commissioner/DG, the Council Presidency, and a negotiating team of the European Parliament led by the respective rapporteur of the committee in charge. It goes without saying that negotiations followed up by a ratification-type adoption in Parliament is a method which does not allow for Parliamentary deliberation and also runs the risk of being intransparent. The institutions are trying to tame that risk by interinstitutional agreements, but the European Parliament should reduce the practice of trilogues and strengthen its own debating and deliberating procedures.

40  The streamlining of ‘sub-statutory’ legislation was seen as a leap forward in the Lisbon Treaty. And indeed, the introduction of Delegated Acts (Delegated Legislation) (Art. 290 TFEU) as well as the regulation of Implementing Acts (Implementing Legislation) (Art. 291 TFEU) are helpful to shape the EU’s legislative framework. In Article 290 TFEU, the Commission is given real power to pass delegated legislation. Such delegated legislation is useful to ‘fine-tune’ any regulatory measure that cannot itself anticipate all future circumstances and requires more detail. The empowerment must be contained in a legislative act and enables the Commission to pass ‘non-legislative acts of general application to supplement or amend certain non-essential elements of the legislative act’. Said article contains provisions to limit and to control delegation. For limitation, delegation only applies to non-essential elements of legislation, and the ‘objectives, content, scope and duration of the delegation of power shall be explicitly defined in the legislative acts’. This limitation is drawn from German constitutional law. For control, the delegation may be revoked either by the European Parliament or the Council and the entry into force of the delegated act may be made subject to the non-objection of either the European Parliament or the Council. Article 290 TFEU thus provides a clear framework for the Commission to regulate administrative details.

41  Article 291 TFEU on implementing acts is less spectacular as it perpetuates the concept of comitology (from the French comité), the control of the Commission’s exercise of implementing powers by committees composed of Member States’ representatives. By means of comitology, the expertise of national bureaucracies is embedded into the rule-making process. Comitology is formalized by means of Regulation 182/2011. Depending on the effects of the implementing act and/or on the respective policy field, an advisory procedure or an examination procedure may be carried out. In the examination procedure, the examination committee may under certain circumstances block the adoption of the implementing act.

42  The institutions are free to choose between delegated or implementing acts in conformity with the conditions provided by the Treaty but they have to justify their choice in the basic legislative act. The main difference between both categories is that only delegated legislation may amend non-essential elements of the legislative act empowering the Commission. The strong Commission power is however weakened by two practical developments: (1) although not foreseen in Article 290 TFEU, the Commission widely seeks advice from expert bodies composed of Member States’ representatives, which runs counter the intended empowerment of the Commission; and (2) agencies are more and more are involved in legislation. The Court backed this development by not considering Article 290 TFEU as a source of exclusive delegation (United Kingdom v Parliament and Council). This, however, is in contradiction with the wording used in Article 288 TFEU. This provision explicitly states that ‘the institutions shall adopt’ regulations etc; ‘bodies, offices and agencies’ are not mentioned.

43  The legal acts enumerated in Article 288 TFEU have long been defined and practised in EU law. The Regulation has general application, ie it is binding and directly applicable in the entire Union. It would have been more than appropriate to designate it as the European law or statute thereby avoiding the ambiguity of the current treaty text which has also consequences for judicial review. Standing of private individuals in an annulment action under Article 263(4) TFEU may be measured by less rigorous criteria in case of a ‘regulatory act which is of direct concern … and does not entail implementing measures’. The Court, in Inuit, rightly found that the concept of a regulatory act does not encompass legislative acts so that a regulation is not the same as a regulatory act, the latter being a typical sub-statutory instrument of rule-making. This uncertainty reveals the deficit of the current wording of the Treaty. Regulations may be legislative acts but also delegated or implementing acts—and in this sense ‘regulatory acts’.

44  The Directive has been used in many fields of law to harmonize national legislations. It is binding only with respects to the results aimed at, but it ‘shall leave to the national authorities the choice of form and methods’ (Art. 288 TFEU). Legislation by directives is therefore legislation in two steps: (1) by the EU legislator and (2) national implementation. Implementation has to be performed within a certain time limit. The CJEU has over time required the Member States to implement directives in a transparent and binding way to avoid uncertainties, in particular if rights are conferred. A particular method to advance implementation is the direct effect of directives. Although provisions in directives need implementation by definition, they may have direct effect if they are self-executing (‘unconditional and sufficiently precise’), the time limit has expired, and if they do not cause a legal duty for citizens. The latter restriction is applied (1) as it is always the Member State and not the citizens failing to implement a directive, (2) legal certainty requires that there is some implementing national norm, and (3) to sanction the non-implementing Member State. This means that there are no obligations of private individuals drawn from the directly applicable provision of a directive. The breach of such a provision cannot result in a criminal offence. Without implementation, directives are also inapplicable in private law relationships. The Court, however, has shown tendencies to modify this settled case law by means of consistent interpretation. In this, the relevant national provision has to be construed in consistency with the provisions of the directive, and this may lead to the same result between private individuals as if the provision had direct effect. The requirement of consistent interpretation is generally applied in an extensive way. Linking it to the realization of fundamental rights, in particular of rights to non-discrimination, it is even used to overcome the settled interpretation of national rules, which leads to interpretation contra legem with respect to national law. The implementation of directives has also been advanced by the Court’s case-law accepting an action on compensation claims against a Member State for damages caused by non- or defective implementation of a directive as a necessary remedy under EU law.

45  Decisions are an instrument for individual cases. They may be general or specify to whom they are addressed. In the latter case, they may entail direct effect.

46  The executive function within the EU is first and foremost exercised by the Commission. However, the Commission has lost much of its executive power in the last roughly 20 years to regulatory agencies (cf above Institutional Pluralization).

47  The CJEU is dealt with in a special entry of this Encyclopedia.

48  Institutional balance is first of all a constitutional principle protecting the interests attached to the various institutions established by the Treaties: the interests of the Union citizens, the interests of the Union as such, and the interests of the Member States. It prevents dominance of a particular interest and preserves a constitutional equilibrium within the EU. In the wording of the Court—used as early as in the aforementioned Meroni judgments—it is deemed to imply an element of judicial protection. As a constitutional principle, it is not equivalent to the classical ‘separation of powers’, but a helpful surrogate. Article 13(2) second sentence TEU adds the principle of mutual sincere cooperation which complements the institutional powers and obliges the institutions to work together in pursuance of the common targets.

49  Every federation, whether a federated State or a federal entity such as the EU, needs an attribution of powers between the federation and the federal sub-entities. This question was outside the main focus in the beginning, as the field of activity of the ECSC and the EEC was rather limited to particular economic tasks. When the activity extended, the need for rules of attribution became obvious.

50  The core principle is the principle of conferral (Art. 5(1) first sentence TEU). Under this principle, ‘the Union shall act only within the limits of the competences conferred upon it by the Member States in the Treaties to attain the objectives set out therein’ (Art. 5(2) TEU). The Treaties categorize the competences of the EU under five headings (listed in Art. 2 TFEU): exclusive competences, shared competences, competences of coordination, the CFSP, and competences of support. Article 3 ff TFEU provide for details within each category. The content of the competence, however, is to be found in the relevant provision of the Treaties itself. A special position is held by the external competence of the Union, ie by its capacity to enter into external relations (Arts 3(2) and 216 TFEU). These articles codify and develop an important line of jurisprudence of the Court and lead to a parallelism of internal and external competences.

51  Uncertainties about the scope of EU competences arise from provisions with particular breadth. Thus, Article 114 TFEU allows ‘for the approximation of the provisions laid down by law, regulation or administrative action in Member States which have as their object the establishment and functioning of the internal market’. This is a very broad competence that is not linked to a particular subject matter. The Court sees a limitation particularly in the test whether the relevant legislative act really aims at eliminating obstacles for the functioning of the internal market or if it has another political target. Article 352 TFEU, an article that somehow codifies the idea of implied powers—but, in practice, has always been treated as something different from the concept of implied powers—enables the institutions to adopt the appropriate measures if the necessary powers are absent to attain one of the objectives of the Treaty. The procedural hurdles set by the Treaties—Commission proposal, unanimous Council decision, and parliamentary consent by a majority of MEPs—avoid the excessive use of this Article.

52  Two principles are established to control the exercise of the competences (Art. 5(3) and (4) TEU). According to the principle of subsidiarity, which applies outside the areas of exclusive EU competence, ‘the Union shall act only if and in so far as the objectives of the proposed action cannot be sufficiently achieved by the Member States, either at central level or at regional and local level, but can rather, by reason of the scale or effects of the proposed action, be better achieved at Union level’. The principle of proportionality, initially developed to limit encroachments of public power upon the citizens, foresees that ‘the content and form of Union action shall not exceed what is necessary to achieve the objectives of the Treaties’. The control of subsidiarity is largely proceduralized in an early warning procedure involving the national parliaments (cf Protocol on the application of the principles of subsidiarity and proportionality). Control by the Court is rare.