1 - Between Crises and Decision-Making

Summary

The chapter identifies the types of crisis exploded in the 20210s and 2020s and the theories on EU decision-making.

1.1 Brussels in Hard Times: Who Decides?

1.1.1 Introduction

How did the European Union (EU) behave in the hard times of the 2010s and 2020s crises? This chapter (Sections 1.1 and 1.2) introduces to the problematiques analysed in detail in the entire book. Since 2009 the EU has seen a sequence of crises that have rocked its very institutional structure. It is noteworthy that 2009 was also the year in which the Lisbon Treaty, the last of the treaties approved, came into force. The idea of the Treaty was to close the long and troubled period of the EU’s institutional consolidation exemplified by the major enlargement in 2004–2007. So, while the Lisbon Treaty thought it had completed the consolidation stage, the crises reopened it. How did the EU institutions perform during the post-2009 crises? To answer that, it is preliminary to identify the features of the EU governance system.

Literature on governance is a growing industry, particularly in the field of EU studies (Börzel 2016; Sabel and Zeitlin 2012; Piattoni 2010). This literature has contributed to our understanding of how a multi-level and complex system, such as the EU, functions. EU governance has been interpreted as ‘a way of governing that does not assume the presence of a traditional, hierarchical government at the helm of the polity’ (Christiansen 2016a: 97, italics in the original). Thus, governance is generally considered to be a horizontal mechanism for dealing with a broad range of problems in which public and private actors interact to reach mutually satisfactory and binding decisions. The opposite of government that is understood as the hierarchical organisation of the decision-making process, where decisions are taken by exclusively public actors controlling the institution at the top of the institutional hierarchy. The distinction between governance and government, although theoretically clear, is more nuanced empirically (see Chapter 5). Most of all, that distinction does not capture the characterising features of the decision-making system that has been institutionalised in the EU.

Here, I will proceed as follows: first, I will introduce the features of the EU governance (discussed more at length in Sections 2.1 and 2.2); second, I will conceptualise the crises of the post-Lisbon Treaty (analysed in detail in Sections 3.1, 3.2, and 4.1), and then I will consider the enlargement’s implications induced by the Russian war (Section 4.2), to discuss their governance by the EU. I will conclude by arguing that the EU has an unresolved problem with executive power.

1.1.2 Monnet and the Crises

The institutionalised area where governance takes place in the EU consists of different decision-making approaches that give form to a composite system (Fabbrini 2015a). According to the 2009 Lisbon Treaty, four institutions participate in the (political) decision-making process, two on both the executive and legislative sides. On the executive side, the European Commission (hereinafter the Commission, consisting of twenty-seven commissioners, including the president of the institution) and the European Council (consisting of the twenty-seven national heads of government, premiers or presidents, plus the president of the institution, the Commission president, and, when international issues are on the agenda, the High Representative of the Union for Foreign Affairs and Security Policy, who is also a vice-president of the Commission, hereinafter the High Representative). On the legislative side, the Council of Ministers (hereinafter the Council, in the various functional compositions of 27 national ministers) and the European Parliament (hereinafter the EP, 705 members elected in the 27 member states, which will become 720 as from 2024), although some configurations of the Council play, in specific policies, an executive role too. Moreover, regulatory agencies and institutions, such as the European Central Bank (ECB), affect the decision-making process, but they will not be considered here because of the technocratic (i.e. non-political) nature of their statute (although their decisions might have political effects). It was with the 1992 Maastricht Treaty that the EU internally institutionalised different decision-making regimes (or pillars), a supranational one for single market regulatory policies and an intergovernmental one for strategic policies, traditionally close to the heart of national sovereignty (core state powers such as security, foreign affairs and defence, home affairs and political asylum, fiscal policy but also health, energy, and military aid policies; Genschel and Jachtenfuchs 2014). The same institutions play different roles in different policies, through different inter-institutional frameworks. The 2009 Lisbon Treaty abolished the pillar structure, but it preserved the decision-making differentiation.

In supranational governance, which was prefigured in in the 1957 Rome Treaties and then clarified with the Single European Act of 1986, the Commission monopolises legislative initiative, while the Council and then (since 1979) the EP have the power whether or not to approve (with differing majorities among them) the proposals (regulations and directives) put forward by the Commission, with the European Council of heads of state and government called on to intervene only when disputes emerge on politically sensitive issues. This decision-making regime was enhanced by various treaties approved after 1992, up to becoming, in the 2009 Lisbon Treaty, the ordinary legislative procedure (Dehousse 2011). Instead, in intergovernmental governance, decisions on core state power policies are rarely of a legislative (but rather political) nature and are taken based on the initiative by one or other national government (rather than by the Commission). The decision-making process is coordinated by the Council and the European Council, with the Commission acting as a secretariat and the EP sidelined (it is informed of the decisions taken but rarely has the chance to approve or reject them) (Bickerton et al. 2015a).

In August 1954, Jean Monnet said something that became an unchallengeable truth in pro-European thinking, that is, ‘Europe will be forged in crises, and will be the sum of the solutions adopted for these crises.’ However, the differentiation in EU governance makes Monnet’s phrase problematic. As Anghel and Jones (2023: 767) noted, ‘Any argument that Europe is forged through crisis is unlikely to tell us much about what Europe is or where it may be headed.’ Considering supranational and intergovernmental approaches, it would be necessary to specify which form of governance is favoured ‘by the solution adopted for the crisis’ in question. This can, in fact, lead to an acceleration of the integration process in either a supranational or intergovernmental direction (Fabbrini and Puetter 2016). Which form of governance came to be favoured by the crises of the 2010s and 2020s?

1.1.3 Conceptualising Crises

Not all crises are equal (Lehne 2022). They differ in terms of their nature, magnitude, and their ‘cognitive construction’ (Schmidt 2015a). Despite reflecting empirical phenomena, crises are generally constructed by constellations of political actors successfully mobilising ideas that fit their own interests or views. In the EU context, the construction of a crisis is conditioned by three main factors. First, by the allocation of treaty competences to deal with the crisis. The crises of the 2010s and 2020s here considered (the sovereign debt crisis, the pandemic crisis, and the energy and military aid crises induced by the Russian war)¹ exploded in policy fields where national governments had statutory pre-eminence over supranational institutions in terms of competence. What Bojar and Kriesi (2023: 431) argued is true, namely ‘Under crisis conditions of high urgency and uncertainty, executive decision-making is generally likely to become the policymaking mode.’ However, because ‘in the multi-level polity of the EU, executive decision-making primarily involves the EU Commission, the European Council, and the governments of the member states’ (Bojar and Kriesi 2023 and Kriesi 2023), the crises examined here prioritised the action of the European Council and national governments rather than the Commission (Zgaga, Capati and Hegedus 2023). Second, by the nature of the crisis, that is, by whether the crisis is due to an internal or external chock (i.e. its source is endogenous or exogeneous) and affects some or all the member states (i.e. its impact is asymmetrical or symmetrical). There is no correlation between the source and the impact of a crisis. Whether the sovereign debt crisis was an endogenous and asymmetric crisis, the pandemic crisis was an exogenous crisis with symmetric effects, the energy crisis had an exogenous origin (the Russian war) but its effects were asymmetric, while the security crisis was exogenous (the Russian war) but its effects were mixed, symmetric regarding the military threat (a threat potentially affecting all the EU member states, but faced through another organisation, NATO) and asymmetric regarding the need to aid military Ukraine (a commitment perceived more by the eastern than the western member states). Different types of crises thus generate different opportunity structures for political action. Third, by the political leadership’s commitment to construct a crisis according to a specific interest or view. If a crisis pertains to the realm of national governments’ competences, that commitment will then be affected by the distribution of national government preferences, as well as by the culture and motivation of the potential leaders (premiers or presidents). In any case, a crisis faced by national leaders will unlikely generate a supranational solution that could jeopardise the power of those national leaders.

The sovereign debt crisis in the first half of the 2010s had a distributive character because it impacted the relations among the member states regarding the costs to be met to manage or resolve it but did not call into question the main paradigm to handle a budgetary crisis. Despite having a distant external origin (in the 2007–2008 global financial crisis), it was constructed as endogenous by the national leaders of creditor states (due to the fiscal profligacy of debtor member states), asymmetric (hitting the southern member states rather than the northern ones), and reinforcing the predominant policy paradigm (each member state is responsible for the state of its public finance, the so-called moral hazard paradigm) (Carstensen and Schmidt 2018). The sovereign debt crisis rocked the Economic and Monetary Union (EMU) or Eurozone, based on a single currency and distinct national budgetary policies, although the latter are highly regulated to ensure their compliance with precise macroeconomic parameters. As Bongardt and Torres put it (2022: 283), the crisis hit ‘an EMU left incomplete in its economic part’. The crisis’ solution led to a strengthening of the regulatory model, in the form of new legal measures and new intergovernmental treaties (outside the EU), to prevent conduct entailing moral hazard by the EMU member states. The dominant policy paradigm was confirmed. It should be the national responsibility to deal with the crisis, also because the debtor member states did not construct an alternative interpretation of the latter. That solution generated a deep division between southern and northern members of the Eurozone (Matthijs and Blyth 2015), with the institutionalisation of reciprocal distrust among them.

The pandemic crisis (exploded in 2020) and the energy and military aid crises (induced by the Russian aggression against Ukraine in 2022) were instead a blow to the structure of the EU and not (as with the sovereign debt crisis) to distributive relations among the member states. For this reason, they can be conceptualised as constitutive crises because they called into question the policy models that have organised the material constitution of the EU in those policy fields. The pandemic’s magnitude showed the inconsistencies in the healthcare model based on the member states’ responsibility to guarantee the protection of their citizens from epidemics (Schmidt 2020a) and the impossibility to rely on national resources for recovering from the latter. The programme of Next Generation EU (NGEU) epitomised a change of paradigm, based, as it was, on EU debt managed by the Commission together with the Council of economic and financial ministers (ECOFIN Council). However, the programme had a temporary character (it should last till June 2026), and the funds derived by the debt were distributed to the member states, although managed by their national governments in accordance with a National plan of recovery and resilience negotiated with the Commission and implemented under the Commission’ supervision. At the end of day, NGEU favoured national governments and not supranational institutions. Also, the war waged by Russia showed the inconsistencies in the growth and defence model adopted by the EU member states with the end of the Cold War. In just one night, the Russian leadership wiped out the efforts by European countries (Germany, in particular) to appease and trade with that country, efforts which were driven by significant economic interests. The post-Cold War approach of ‘peace through trade’ had enabled Germany to enjoy low-cost energy with which to support its national industry, thus making their products competitive on international markets (specifically the Chinese market). Despite the Russian annexation of Crimea and some eastern areas of Ukraine in 2014, Germany (but also Italy and the countries connected to German industries’ chain of values) had continued to rely in industrial terms on two authoritarian regimes, Russia for energy and China for markets (Dempsey 2022). At the same time, the military security of Europeans continued to be guaranteed by the US commitment to NATO (an example of European free riding).

However, contrary to the pandemic that led to a new paradigm of policy through NGEU (a common threat should be dealt in common), the Russian war did not change national preferences (favouring national and not European answers to its energy and security implications). Putin aggression was faced with the policy of economic sanctions against Russia and the reduction of the energy’s dependence on Russia, a policy that affected the EU member states differently. At the same time, the answer to the military side of the aggression was left to NATO, with the EU member states limiting themselves to help Ukraine through transfer of finance and weapons, a commitment that, too, was differently interpreted by them. The Russian military aggressiveness represented certainly a collective challenge (threatening potentially all the member states); however, it was constructed as a challenge for NATO more than the EU. After all, the EU had no military defence system of its own, despite the rhetoric about its strategic autonomy, depending completely on the Americans through NATO. The EU found itself without a European defence industry too, a sector that was fragmented owing to jealousy among the various member states. Military aid to Ukraine was thus the only policy the EU could pursue through coordination of national governments in the European Council; yet also, this policy approach showed differences in interests, resources, and views among them.

At the end of 2023, national governments agreed (with the constant opposition of the Hungarian government) on twelve sanction packages, promoted by the High Representative, as well as on helping Ukraine militarily through the intergovernmental programme of the European Peace Facility and then the supranational programme of the Macro-Financial Assistance Instrument (MFA+) (only for 2023, with features like the NGEU). Yet, the policy coordination model, adopted for reaching those goals, could not go further, not only because no national leader emerged (as during the pandemic) for proposing a collective solution to the energy and security transition problems (as it was NGEU) but also because national leaders were elected for making primarily the interests of their domestic voters end not those of the EU as such. If the paradigm of ‘national responsibility first’ was confirmed by the sovereign debt crisis, it was challenged by the pandemic, and the energy and security consequences of the Russian invasion of Ukraine brought it back again. Notwithstanding the different source and impact of the crises, intergovernmentalism (in its various sub-types, tight as in the fiscal policy or loose as in the energy policy) remained the predominant governance model for managing both distributive and constitutive crises. In all those crises, national governments claimed pre-eminence in competence ahead of supranational institutions (Ferrara and Kriesi 2022), although they needed the Commission’s support for delineating their proposed solutions. In this context, the European Council acted as the institution where the medium-term solutions to the crises were devised, not only as the institution for managing the crises in the short term. Let us now introduce the governance of the crises of the 2010s and 2020s.

1.1.4 Taking Decisions in Hard Times

In the sovereign debt crisis, successfully constructed as endogenous and asymmetric, the European Council (and the Euro Summit of the national leaders of the member states of the Eurozone) had the strict monopoly over decisions, with the support of the economy and finance ministers of the Eurozone or Eurogroup. Within the latter, the leadership (in terms of resources and communication) was exercised by the northern creditor states, Germany in particular. The solution to the crisis promoted by the national government leaders of the north led to the strengthening of intergovernmental governance, also thanks to the intergovernmental treaties agreed outside the EU. However, intergovernmental governance created more problems than solutions (see Section 3.1). The Euro Summit/European Council, which functions according to a criterion of unanimity, cannot handle crises that need immediate and efficient responses (Fossum 2020). During the sovereign debt crisis, the European Council was indeed criticised for decisions that were ‘too little and too late’. When a decision implies the unequal distribution of costs and benefits, then the deliberative nature of intergovernmental governance disappears, to give way to more Weberian power relations (in the case of a financial crisis, of creditor countries over debtor countries). Once constructed as a distributive crisis, the intergovernmental decision-making process ended up in generating costs for some and benefits for other member states. An outcome that, without the EP allowed to take part in the decision-making process, is inevitably perceived as illegitimate by those who pay the costs (the citizens of the debtor countries). Indeed, the institutional solution adopted for the sovereign debt crisis (defined as unconstrained intergovernmentalism because it made tighter the principle of ‘national responsibility first’) triggered a populist reaction in almost all the debtor countries, delegitimising the Eurozone system overall (Hopkins 2020). Populism has become, since then, a permanent feature of European democracies (Mény and Kermer 2021).

With the pandemic, successfully constructed as exogenous and symmetrical, an intergovernmental solution struggled to take hold, although the European Council was the centre of the decision-making process (Fabbrini 2023b). Moreover, some national (the German chancellor) and European leaders had personally experienced the destabilising effects of that approach (an example of policy learning). As no one could be held responsible for the pandemic, the crisis was framed as a common threat that required a common (not to confound with European) response. That response arrived through interstate divisions, implying different narratives about the crisis, between coalitions of member states (see Section 3.2), as indeed happened also in established federations such as the United States of America or USA (Fiorina 2023), where however the interstate division was ideological rather than territorial. Because of the magnitude of the pandemic, and the conflict of interests between the member states, the European Council had to solicit (or to accept) a more active role by the Commission. Due to the rising costs for vaccines generated by competition among states to buy them, the Commission had to step in as the sole agency for their purchase, thus lowering their cost. Given the interstate divisions generated by the need to support the various national economies, it was necessary to acquire new resources for the recovery and resilience of national economies through the NGEU. This is a programme consisting of loans and grants (guaranteed by the budget of the EU and those of its members states as well as by own new resources, so far only the plastic tax) managed by the Commission and the Council, under the supervision of the European Council yet deprived of the power of veto. The supranationalism of the NGEU was, however, constrained in terms of time (the programme will end in 2026), institutional scope (the EP had no decision-making role to play), and logic (the resources acquired are distributed to member states and not used autonomously by supranational institutions). One might say that it is a supranationalism at disposal of national governments, who are the owners of the programme. The latter is co-managed by the Commission with the national governments, where the Commission has not the power to pursue its own policies. With the pandemic, the EU was given an ad hoc fiscal capacity (Fabbrini 2022), however temporally and dependent on national governments preferences. Not exactly the capacity for dealing with new challenges.

Challenges that arose with the Russia’s aggression against Ukraine. For facing that aggression, the European Council and national governments claimed their decision-making competence, although they had to rely on action by the Commission (and the High Representative) to advance immediate common answers. Through the High Representative’s initiative, the European Council has approved, at the end of 2023, twelve packages of economic sanctions against Russia, despite repeated veto’s threats by the Hungarian government delaying their implementation. However, the need to reduce the dependence on Russian gas led to divisions on how (and whether) to control its price, since national policies on energy differed significantly. Taking advantage of the temporary suspension of the regulation that prohibits state aid,² some national governments intervened with policies to support companies and citizens in the energy transition. On 9 November 2022, under the Temporary Crisis Framework, for a total amount of approximately 253 billion euros authorised by the Commission as state aid, two member states accounted for roughly the 80 per cent of the requests (France for 160 billion and Germany for 36 billion euros). Thus, the Temporary Crisis Framework accentuated the differences between countries that had fiscal space for helping their firms and families and other countries restricted by high public debt. The intergovernmental model of national policy coordination has thus further fragmented the EU energy policy.

Also, the military assistance for Ukraine ended up having distorting effects. Some member states sent old munitions to Kyiv but then asked to be reimbursed as if they were new (Finland claimed 100 per cent of the reimbursement based on new purchase prices, Latvia claimed 99 per cent under those terms, Lithuania 93 per cent, Estonia 91 per cent, France 71 per cent, and Sweden 26 per cent). Facing competition among national governments to buy new armaments, with the massive rise in the costs of military materiel, the European Council had to ask the Commission to become a common procurement agency for those armaments, thus negotiating lower costs with the companies that make them, with the help of the European Defence Agency headed by the High Representative. The financial resistance, due to their lack of resources, of the member states finally led the Commission to promote a new programme, MFA+, to aid Ukraine militarily but only for 2023, funded by European debt in the same way as the NGEU. Finally, the Commission’s attempt to set up a Ukrainian Facility consisting of 54.6 billion euro over four years (2024–2027), out of the mid-term revision of the multiannual financial framework2021–2027, was blocked by the Hungarian veto during the European Council meeting of 14–15 December 2023, since budget revision requires the unanimity of the European Council for being implemented. With the war continuing, differences among member states regarding the military aid to Ukraine increased, with some national governments (as Italy, France, and Spain) transferring to Ukraine much less resources and weapons they were committed to give (see Section 4.1).

Through the ‘policy coordination’ model (see Section 2.1), national governments looked for national solutions to their energy and security needs, also because they were differently affected by the energy crisis and by the necessity to help Ukraine (a necessity perceived more in the eastern than in the western member states) (see Section 4.1). While the sovereign debt crisis further strengthened the intergovernmental cage and the Russian war increased the reliance on national capabilities, the pandemic crisis opened the possibility of national governments to resort to a European debt, a significant step forward for the EU, although the debt was used by national governments and not supranational institutions. Table 1.1 summarises the features of the crises and their institutional outcome (that I will then analyse more in detail in Sections 3.1, 3.2, and 4.1, while Section 4.2 will discuss the enlargement process reopened by the Russian war). In all the crises, the European Council emerged as the centre of the EU decision-making process and the solutions it offered to the crises further strengthened its centralising role. Going back to Jean Monnet, one might thus argue that, if it is true that ‘Europe will be forged in crises and will be the sum of the solutions adopted for these crises’, then the latter have strengthened the intergovernmental rather than the supranational component of the EU.

Table 1.1 The crises of the 2010s and 2020s

Nature of the crisis	Sovereign-debt crisis	Pandemic crisis	Russian war (energy)	Russian war (military aid)
Source	Endogenous debtor member states	Exogenous external shock	Exogeneous external shock	Exogeneous external threat
Impact	Asymmetrical some states affected more than others	Symmetrical all states are affected	Asymmetrical some states affected more than others	Symmetrical (military threat) Asymmetrical (military aid)
Outcome	Unconstrained intergovernmentalism – ESM	Constrained supranationalism- NGEU	National policy coordination –sanction packages	National policy coordination – MFA+
Implications	Division between northern and southern states of the Eurozone	Division between three groups of states	Division between energy-dependent and energy independent states	Division between states with different geopolitical positions

Above all, the Russian war showed the unpreparedness of the EU and its member states to face a military threat. Theoretically, the Russian war represented the threat that, in the Riker’s tradition (Riker 1964), could incentivise the centralisation of fiscal and military capabilities, as shown by the process of state-building (Kelemen and McNamara 2022). However, the Russian war was interpreted, by the generality of national governments, as a threat that NATO, and not the EU, must face. Indeed, the Russian war led to the strengthening of NATO, also through its enlargement to Sweden and Finland, traditionally neutral countries. The Russian war highlighted the lack of European defence, despite the Permanent Structured Cooperation on Security and Defence, envisaged by Articles 42(6) and 46 as well as by Protocol no. 10 of the Treaty on European Union. Just consider that, one month after the Russian invasion, the EU approved (March 2022) a Strategic Compass, proposed by the High Representative, which envisages the mobilisation of a European rapid deployment force of just 5,000 soldiers. National governments considered that it was possible to continue to free-ride on American military protection to guarantee the security of Ukrainian sovereignty. When, in the following months, various national governments finally set themselves the goal of investing in defence, they invested in national and not EU defence, with the result of exacerbating the asymmetries among the various national defence capabilities. Think of the approval (in June 2022) by the German Bundestag of 100 billion euros to be spent on defence. Certainly, it was a zeitenwende or watershed in German defence policy (which implied emending the Fundamental Law) (Scholz 2023), but it was not a step towards a supranational EU defence system. Indeed, it is likely that Germany’s asymmetric rearmament will arouse inevitable concerns in other EU member states, which will be hard to control through intergovernmental coordination. In short, the crises highlighted the incongruence of a union ‘governed’ by twenty-seven national governments. Intergovernmental governance cannot identify a common interest in dealing with a crisis. After all, national government leaders were elected to promote the interests of their own country and not those of the EU.

1.1.5 Conclusion

As soon as the Lisbon Treaty came into force in 2009, the EU had to face a series of crises which shook the institutional structure formalised by that Treaty. Contrary to the expectations of its internal and external opponents, the EU survived those crises, showing that it has sufficient institutional resilience. However, contrary to the opinion of its internal and external supporters, the EU showed that it was equipped to react rather than to act (a debate well analysed by Tsoukalis 2023). After exploding in policy fields with a pre-eminent competence of national governments (such as fiscal, health, military, industrial and security policies, but the same is true also for the migration and asylum policies), those crises were initially addressed through the European Council. The role of the European Council/Euro Summit led to an unconstrained intergovernmental solution to the sovereign debt crisis, it was crucial for finding a constrained supranational solution to the pandemic crisis, and it was central in the form of policy coordination for answering the energy and security crises induced by the Russian invasion of Ukraine. However, a decision-making process based on the consent of twenty-seven national leaders proved to be slow and cumbersome. Particularly during international emergencies, an inevitable tension came to emerge between the European Council and the Commission, or better between their respective presidents, aiming to establish ‘who’s in charge?’ or ‘who represents Europe?’, with the outcome of increasing confusion within and outside the EU.³

Moreover, if the European Commission is controlled by the bicameral legislature (the Council and the EP), the European Council is instead not subject to checks and balances at the level at which it acts (see Section 2.2). Its members take decisions in an accountability vacuum. In fact, despite individual national leaders having the confidence of their respective parliaments or electorates, the European Council as an executive institution is not controlled by a legislative institution (the EP) operating at the same level. The EU’s executive deficit consists of the absence of a single executive institution endowed with the capabilities and legitimacy to act, in turn controlled by a bicameral legislature legitimated by European citizens and by member states. The EU must not become a state, but it cannot do without a unitary, effective, and legitimate executive able to guarantee its external security and its internal functioning. The representational deficit in the EU has long been discussed, yet the 2010s and 2020s crises have shown that the EU has a governability deficit. Brussels has an executive power’s problem in dealing with hard times (Fabbrini 2023a). How has executive power been conceptualised by the main theories of European integration?

1.2 Theories of EU Decision-Making

1.2.1 Introduction

Executive power is one of the least investigated areas of the EU, probably because executive power is one of the least defined and most ambiguous components in the EU institutional system. Nothing new under the sun, if one considers that executive power is the most controversial power in any aggregation of previously independent states. From a comparative perspective, however, new under the sun is the sui generis conceptualisation of executive power in the EU. According to Sartori (1991: 247), ‘parochialism refers … to single-country studies in vacuo, that purely and simply ignores the categories established by general theories and/or by comparative frameworks of analysis, and thereby increasingly invent, on the spur of the moment, an ad hoc, self-tailored terminology’. To free the investigation of executive power from the constraints of an ad hoc approach, it is necessary to develop an institutional analysis based on the criteria and concepts used by the literature on comparative federalism (Fossum and Laycock 2021; Fossum and Jachtenfuchs 2017; Kelemen 2019; Fabbrini 2015a; Stepan 1999; Sbragia 1992), distinguishing between the cases that are most similar to the EU (federations by aggregation of previously independent states) and the most dissimilar (federations by disaggregation of a previously unitary/centralised state) (on the strategies of comparison, see Sartori 2009a), where executive power is the dependent variable to identify.

Here, I will proceed as follows. First, I will discuss the most influential studies on post-Maastricht EU executive power, to show their ad hoc understanding of the latter’s features and accountability. Second, I will investigate executive power features and accountability patterns in democratic federal systems, comparing coming-together federations, where the political accountability of executive power is institutional, with holding-together federations, where such accountability is electoral. The distinction between the two types of federation is crucial in appropriately conceptualising the EU. I will use the adjective ‘federalist’ (rather than ‘federal’) to characterise the specific logic of federations by aggregation (such as the USA, the rationale for the foundation of which was in fact explained in The Federalist, now Beard 1948, and Switzerland) from the logic of federations by disaggregation (such as Germany, Austria, Belgium, Australia, and Canada – Canada with some caveats). The former, and not the latter, are the most similar cases to the EU (I developed this argument in Fabbrini 2020a).

1.2.2 Integration Theories and Executive Power

Investigation of EU executive power was not the analytical concern of the first generation of EU studies. The mainstream theories of integration (neo-functionalism and liberal intergovernmentalism) focused on the process of European integration rather than on its institutional outcomes. For such theories, institutions were the contingent outcome of either the functional solution to a policy problem or of the need to guarantee the credibility of the commitments entered by governments in their negotiations. According to Schmitter and his colleagues (Nieman, Lefkofridi, and Schmitter 2019: 45), neo-functionalism has always considered integration ‘to be a process rather than an outcome or an end state’, a process involving ‘the creation and role expansion of distinctive regional institutions’. According to Börzel, (2016: 12), ‘the EU’s governance has evolved over time developing different varieties of inter- and trans-governmental negotiation and regulatory competition in the shadow of supranational hierarchy’. For scholars adopting this approach (Dehousse 2011), the Commission, allied with the European Court of Justice (ECJ) (formally called the Court of Justice of the EU), constitutes the closer approximation to an executive power driving the integration process. According to liberal intergovernmentalism (Moravcsik 1998), instead, that considers the EU to be an organisation of international cooperation, the existence of an executive institution distinct from member state executives is unnecessary. If national governments ‘are unsatisfied with particular formal institutions, [they] can and do replace them with informal norms, opt-out clauses, exceptions and multi-track arrangements. Governments retain a surprising ability to design (and redesign) EU norms and institutions so that they continue to serve broader national interests’ (Moravcsik and Schimmelfennig 2019: 80). National leaders did that through their regular summits and the European Council’s informal (before their formalisation by the 2009 Lisbon Treaty) and then formal meetings. In fact, ‘the European Council, where member states act by direct consensus, now initiates both every day and long-term policies (…)’ (Moravcsik and Schimmelfennig 2019: 80). A ‘processual’ approach to EU governance made comparison with other multi-level and federal systems uncertain or unlikely, further justifying the assumption of the EU as an ad hoc system.

With the full institutionalisation of the EU, in the post-Maastricht period, a second generation of EU studies came to focus on the EU as a political system and no longer as the form of an ongoing integration process. Moving from a process-oriented to an institution-oriented approach made it possible to better investigate the EU decision-making structure. Two approaches can be identified. The first group consists of scholars from different disciplinary backgrounds but sharing a common view on the EU decision-making process (I call it the executive order view). The second group consists of scholars engaged in promoting a new intergovernmental approach to the EU decision-making process (I call it the intergovernmental order view). Let us start with the work by Hix (1999) and Curtin (2009), who conceptualised an EU executive order (to use Trondal 2010) as constituted by a relatively stable framework within which distinct institutions perform the same functions at the same time. As Hix (1998: 41, italics in the text) argued, ‘executive, legislative and judicial powers are exercised “jointly” by the EU institutions. Instead of a classic “organic” separation of powers into three different institutions, there is a “functional” separation of powers across several institutions: with hierarchies within each governmental power.’ As all European nation states have developed democratically from the principle of putting popular sovereignty in the hands of a parliament, starting with the British and Scandinavian experiences, the EU, too, should strengthen the decision-making role of the EP in affecting the composition and the platform of the European executive (the Commission). This outcome has continued to be contested, however. In the year when the Lisbon Treaty came into force, Curtin (2009: 14) could write that executive power ‘consists of various bits and pieces that have been cobbled together across a spectrum of institutions, sub-actors and policy areas’.

Executive power is thus exercised by several institutions, with ‘a number of “satellite” actors who have, in a variety of ways, had executive powers delegated to them but whose role is even more “in shadow”’ (Curtin 2009: 14). In this framework, certainly, the Commission represents ‘a new and distinctive executive centre at the European level, outside of the intergovernmental locus constituted by the Council’ (Curtin and Egeberg 2008: 639), an executive centre ‘organized much like a domestic government, with a core executive (the College of Commissioners) focusing on the political tasks …’ (Hix 2005: 40–41). However, the Commission could not keep for itself the power of an executive. As Curtin (2014: 7) noted, ‘empirical evidence points to a “progressive erosion” of the Commission’s power of initiative, and the European Council’s detailed setting of the legislative agenda is pronounced. The European Council is … a type of default “crisis manager” of the EU (…). Given that the EU has found itself in a more or less perpetual crisis in recent years this key role accentuates a strong executive power’. A sort of division of labour took place during the early 2010s crises, where ‘the European Council sets the agenda and directs the law-making institutions, the Commission proposes the content of far-reaching legislation, ensures its implementation and negotiates international agreements…’ (Curtin 2014: 1). Although the European Council increased its decision-making role, it could not keep for itself either the power of an executive. The Commission continues to remain the intellectual and technical engine of EU decisions, through its capacity for policy design and implementation, surrounded by a panoply of agencies contributing to the decision-making process. Thus, both Curtin and Hix share the view that the EU has an executive order (rather than an executive power) based on the overlapping of different institutions, performing the same executive functions in different times and contexts, an executive order that has supported the supranational integration of the EU. In this multifaceted executive order, accountability is necessarily multifaceted and sui generis. There is no room for political accountability as such.

For another group of scholars (Bickerton et al. 2015a), the early 2010s crises instead showed the undisputed affirmation of an intergovernmental order within the EU decision-making framework. According to these scholars, the European Council has become the decision-making centre of EU politics, also thanks to the Commission’s complicity (Bickerton et al. 2015b: 5). For the promoters of this new intergovernmentalism approach, the post-Maastricht EU has set up new institutions, distinct from the supranational framework, that ‘enjoy considerable autonomy by way of executive and legislative power and have a degree of control over their own resources (…and through which) Member States pursue more integration but stubbornly resist further supranationalism’ (Bickerton et al. 2015b: 3). Member states continue to be crucial in the functioning of the EU, but they are no longer the nation states of liberal intergovernmental theory, nor are their institutionalised interactions confined to intergovernmental conferences or summits. Now, they are supported by a permanent structure, they coordinate in a regularised manner, and they are headed by permanent presidents (in the case of the European Council, the Eurogroup and the Foreign Affairs Council). The European Council has become the catalyst for the integration process (Puetter 2015), with the Commission ‘taking a pragmatic and constructive approach’ to adapting to a more constrained executive role in ‘the new area of EU activity’ (Bickerton et al. 2015c: 308), which meant an increase in its policy implementation capability. Also, in the old areas of EU activity, argued Puetter (2014: 1), ‘the Commission’s traditional right of initiative is either formally or de facto limited (…) (while) the European Council has obtained a leading role in policymaking’. In this decision-making context, political accountability is considered to take place within the intergovernmental institutions, through the deliberative process that incentivises their members to respond to everyone’s preferences and requests (Puetter 2012). For both schools of intergovernmentalism, there is no need to respect the criterion of distinction between institutions and functions as in democratic governmental systems. The fundamental distinction (either institutional or functional) in each system of government between the executive and the legislative branches (and between these and the judiciary) tends to blur in the intergovernmental views of the EU. The intergovernmental paradigms do not concern themselves with whether the EU risks being organised according to the (undemocratic) principle of a confusion of powers. If the confusion of powers concerns the EU but not its member states, then it will remain a non-issue for both paradigms.

The analytical move from focusing on the integration process to focusing on the latter’s institutional outcomes has allowed a better understanding of the EU’s decision-making structure. For Hix and Curtin, what has emerged is an executive order where several institutions perform similar functions in different times and contexts. For Bickerton, Hodson and Puetter, instead, the decision-making centrality acquired by the European Council has overshadowed the other institutions (including the Commission). Both groups of scholars acknowledge the existence of an EU executive order rather than executive power, a differentiated order with ad hoc political accountability patterns. However, considering the EU a sui generis organisation does not help conceptualise the properties of the executive power and the accountability patterns of its top decision-makers. It is necessary to enlarge the perspective by comparing the EU with organisations that have a federal form. This is the task of the next section.

1.2.3 Comparative Federalism and Executive Power

Comparative politics literature (Lijphart 1999) is useful for conceptualising the features and accountability of executive power in democracies. However, the basic units of comparison remain the nation states. Federations are understood (Beramendi 2007: 760) as variants of nation states, which differ from unitary nation states solely in terms of the organisation of vertical relations between the centre and the territorial units. Indeed, analysis of comparative federations shows that not all of them can be considered variants of nation states. Based on the distinction (used in comparative federalism studies) between holding-together and coming-together federations (Stepan 1999) or between federations by disaggregation and federations by aggregation (Sbragia 1992), one might argue that the first, but not the second, can be considered a variant of nation states. Thus, it is not the generic comparison with federal systems that might help to better understand the EU’s executive power, but the specific comparison with those federal systems which have emerged from the aggregation of previously independent states, as has been the case of the EU. According to Sartori (1991: 244), since comparison is a method for checking hypotheses, adopting the most similar strategy makes comparative checking more effective in establishing ‘whether generalizations hold across the cases to which they apply’.

Horizontally, holding-together federations (which are the most common type) are organised through fusion of powers relations between the executive and (the majority of) the popular chamber, as is proper of parliamentary government in unitary nation states. Certainly, in these federations, the decision-making power of the fused parliament–cabinet majority is mitigated by the influential role exercised by the higher chamber, particularly in policy fields (such as budgetary issues) of high interest to the territorial units. The upper chamber may represent the territorial units of the federation in different forms, either through their governments or through representatives directly elected by their voters. In Europe, Germany, Belgium, and Austria too (notwithstanding the direct election of the latter Republic’s president) are cases of parliamentary federations where the executive power is connected through a confidence relation with the lower chamber’s majority. Outside of Europe, among OECD countries, Canada and Australia are cases of established parliamentary federations.

These parliamentary federations (with the partial exception of Canada, Fossum, and Menendez 2011) have historically emerged from the disaggregation of a previously unitary state or imperial dominion, with the centre controlling the administrative resources and political personnel to preside over the process of federal decentralisation. This is true also for post-Second World War Germany, the federalism of which is the outcome of the Allies’ decision to decentralise the powers of the Third Reich’s highly centralised state rather than the autonomous decision of independent Länder to aggregate to form a federation (Hueglin and Fenna 2015: Ch. 5; Jefferey and Savigear 1991). The territorial borders of previous Länder were recognised in only a few cases, many new Länder were formed by either aggregating or separating previous Länder (on this, see Section 5.1). The artificial definition of several German Länder makes the argument that post-Second World War German federalism is the outcome of the aggregation of previously independent sovereign states implausible, but it helps understand why post-Second World War German federalism came to be organised around a parliamentary government model.

The sustainability of parliamentary federations is due to many factors, but certainly, the homogeneity of the national political culture has historically played a positive role. A homogeneous national political culture helps contain the domestic political conflict within the limits of distributional issues and not territorial identity cleavages. According to Scharpf (2008: 510), post-Second World War Germany is a ‘federal state with a unitary political culture. It is a federal state with parliamentary governments at the national level (because ed. added) there are no politically salient territorial cleavages defined by ethnic, linguistic, or religious divisions, and no popular demands for regional autonomy’. The 1990 annexation, by the Federal Republic of Germany (Bundesrepublik Deutschland) of the previous German Democratic Republic (Deutsche Demokratische Republik), in the form of five new Länder plus East Berlin, incentivised political inhomogeneity between the old and the new Länder, but none of the new Länder has ever claimed a distinct national identity from the old Länder. When, in parliamentary federations, the political cleavage assumes a territorial, identity-based character (as in Belgium, with the Flemish community’s request for autonomy or in Canada, with Quebec province’s aspiration to national independence), then their sustainability (and legitimacy) is called into question. This is true also for those quasi-federal parliamentary systems (such as Spain or the United Kingdom), where centrifugal pressures derive from territorially based communities that identify themselves as distinct nations (such as Scotland or Catalonia). Parliamentary federal democracies have displayed a high level of complexity in their administration and policymaking machinery (one has only to think of the German model of cooperative federalism, Scharpf 1988). Notwithstanding that complexity, in those federations, the government has continued to be the institution which acts as the ultimate decision-maker, and the popular chamber, whose political confidence is necessary for the government to operate, has continued to embody the substance and symbolism of unified national sovereignty. In parliamentary federations, the government’s political accountability is guaranteed by the periodic elections of the popular chamber (electoral accountability), since the voters have the possibility to express a judgement on the party/parties constituting the government. At the same time, between elections, accountability of the government is promoted by a plurality of media and political actors, starting with the parliamentary opposition.

Parliamentary federations, however, represent only one species (although the most common) of the genus of democratic federations. There is also another species of federation, the one that emerged from previously independent states that decided (mainly for security reasons, Riker 1964) to move from a confederal to a federal pact (precisely, coming-together federations). The USA and Switzerland are the only two cases of successful establishment of democratic federations by aggregation (Kelemen 2014). Contrary to holding-together federations, they have adopted forms (although different between the two) of multiple separation of powers, both vertically and horizontally. Through multiple separation of powers, coming-together federations have sought to divide sovereignty, first between the federated states/cantons and the federal centre and then between the institutions of the latter. The differences between states or cantons (in terms of territorial identity and demographic size) have precluded the adoption of a fusion of powers model, at the horizontal level, with its celebration of the primacy (in terms of democratic legitimacy and the governmental role) of the popular chamber over the other institutions. In coming-together federations, the two chambers have the same legislative power. The exclusive legislative-governmental centrality of the popular chamber, if not supported by a shared national culture and a reasonable demographic balance between the territorial units, might degenerate into the dominance of the larger over the smaller states (Elazar 1987). Horizontal separation of powers aims to prevent this outcome through the fragmentation of decision-making power across more than one institution (it is the Madisonian democracy conceptualised by Dahl 1956, now 2006). In a separation of powers system, the executive and the legislature are not connected by a relationship based on political confidence, but they are reciprocally independent in their operation, although they share decision-making responsibility (Vile 1967). The reciprocal independence of the executive and the legislature means that the former cannot be dissolved by the latter, nor does the latter automatically dissolve with the resignation of the former. Their electoral mandate is constitutionally fixed and cannot be shortened (unlike in parliamentary systems when special circumstances require that). That reciprocal independence is further strengthened by the prohibition, for an individual politician, to hold (unlike in parliamentary federations) an executive office and a legislative seat at the same time. Executive and legislature are ‘separated institutions sharing governmental power’ (Neustadt 1990: 29).

In these federations, voters elect the representatives of the separated institutions, not a political majority as in parliamentary systems (Coultrap 1999). In neither of the two democratically established coming-together federations is the executive directly elected by the voters. In the USA, the president is indirectly elected through an electoral college (which dissolves after the president’s election). The existence of the electoral college has certainly not prevented presidential elections acquiring a popular character (Polsby, Wildavsky, and Hopkins 2011). The development of the two-party system has led to the transformation of the electoral college’s indirect election of the president into the popular (but not direct) election of the latter (Polsby 1983). However, this popular election has had to reckon with the institutional logic of the system, as in those cases of presidential candidates (such as Hillary Clinton in 2016 or Albert Gore in 2000) who won an absolute majority of popular votes but not of state electoral college votes (thus losing the presidential election). In Switzerland (Vattel 2019; Trechsel and Kriesi 2008), the executive is a collegial institution (a directoire, or federal council, of seven members) elected by the two chambers of the federal legislature (the Federal Assembly is made up of the National Council and the Council of States), but (once elected) no longer dependent on the latter’s confidence.

Although not accountable to voters, the executive, either monocratic or collegial but unitary in both cases, is instead politically accountable to the legislature (institutional accountability). In coming-together federations, the role of the legislature is not to form a government (as in holding-together federations), but to control the executive (for the distinction between the parliamentary and the congressional model of legislature, see Kreppel 2011). Executive accountability is guaranteed by the sanctioning powers of an independent and co-equal legislature. This does not mean that the executive has always accepted to be controlled by the legislature (Kriner and Schikler 2016), nor that the legislature has always had the political will to exercise its control over the executive. Particularly in the USA, presidents have regularly tried to escape congressional supervision (using, for instance, their role as commander-in-chief in security and foreign policies, Fisher 2013) or tried to impose their choices on their congressional party (Goldgeier 2018). Presidential attempts to bend the legislature to his own will reached alarming heights during the Trump mandate (2017–2020). Trump systematically defied the checking role of the Congress up to the point, at the end of his mandate, of challenging the Congress’s constitutional duty to certify the votes of the state electoral colleges for the presidential elections, no less asking its supporters to invade Capitol Hill on 6 January 2001.

Thus, in holding-together federations, executive power coincides with the government. Its top political decision-makers consist of the leaders of the party/parties holding ministerial positions that will have to account to voters at the next parliamentary election. In coming-together federations, instead, executive power is a branch of a separated government, whose political accountability has an institutional character because it consists of the control and scrutiny of executive officials by the two chambers of the legislature. In both cases, the executive is a unitary institution. Certainly, the two types of federation express distinct patterns of the political accountability of executive power since they are based on different constitutional frameworks.

1.2.4 Conclusion

The chapter has shown the haziness of the theoretical debate on EU executive power because it is either lacking a comparative reference or uses comparison as an evocation. Since the EU is a union of states by aggregation, I have pursued a similar cases’ strategy for conceptualising its executive power. Yet, the theoretical debate has been, and continues to be, encumbered by the idea of ‘EU exceptionalism’, which has normalised the contradictory coexistence of the European Council and the Commission as components of the executive power. The EU has thus come to acquire a system of governance constructed according to specific ad hoc procedures, which need not necessarily consider the criteria of effectiveness and legitimacy that inspire the functioning of the democratic system of federations by aggregation. This idea has contributed to justifying the sui generis narrative of the EU, based on the assumption that, due to ‘its unique institutional nature’, the EU is an exceptional, unprecedented political system (Orbie 2009: 2), ‘different from pre-existing political forms (because of) its historical context, hybrid polity and political-legal constitution’ (Manners 2002: 240–242). The exceptionalist idea has continued to be particularly common among officials and politicians operating in Brussels. On 27 May 2020, the President of the European Commission, Ursula von der Leyen, tweeted that the EU is a ‘Union of peace and prosperity, without peer or precedent anywhere in the world’. Ten years before, in a speech given at Humboldt University of Berlin on 9 May 2011, Michel Barnier, then European commissioner for internal market and services, asserted that the EU ‘is unique in history and in the world’.

The chapter has called into question this view, bringing in the literature on executive power, its features, and accountability patterns, in federations, distinguishing between those emerging from the aggregation of previously independent states and those emerging from the opposite process of disaggregation of a previously unitary state. The political world is larger than the EU. ‘He who knows only one country knows none’ (Sartori 1991: 245). With this theoretical and comparative background, it is now time to reconstruct the process of institutionalisation of executive power in the EU and the formation of a dual structure, apparently unknown to much of the literature above considered. A dual structure that can help to understand the EU unsatisfactory performance during the crises identified in Section 1.1.