Democratic legitimacy in the post-crisis EMU

ABSTRACT

This paper examines the democratic legitimacy of Economic and Monetary Union (EMU) since the international financial crisis hit the euro area. From its inception, EMU has been marked by an asymmetry as its monetary pillar relied on output legitimacy while its economic pillar relied essentially on input legitimacy at the national level. The crisis severely challenged EMU’s output legitimacy, which led to the establishment of new European-level institutions. We analyse the European Stability Mechanism, the European Semester, and Banking Union to take stock of their powers and the ways that these are balanced by mechanisms of legitimacy. Our main finding is that the intergovernmental and output-oriented approach that originally informed the legitimacy of EMU has remained prevalent after the crisis. However, as the three domains – in varying degrees – address questions that are essentially political, we argue that the channels for input legitimacy at the European level remain deficient.

KEYWORDS:

• Economic and Monetary Union
• legitimacy
• European Stability Mechanism
• European Semester
• Banking Union

1. Introduction

In its original conception, the legitimacy of Economic and Monetary Union (EMU) was primarily a function of the output benefits that monetary stability would bring. The European System of Central Banks was to work towards the objective of price stability and to do so under conditions of political independence (Committee for the Study of EMU 1989, 21f.). At the same time, macroeconomic policy would remain mostly in the hands of the member states themselves (ibid, 19) and, hence, be subject to democratic controls at the national level. Nevertheless, as Verdun and Christiansen (2000) already argued at the time that it was formally established, EMU was essentially a political project by economic means. Rather than being able to rely on a pre-given integrated democratic polity, EMU was meant to contribute to the emergence of such a polity. Notably, however, all of that promise and, indeed, its legitimacy hinged on EMU delivering tangible economic benefits; ‘once EMU provides economic effects, its institutions will gain credibility and legitimacy’ (169).

While after EMU’s first ten years, the verdict on its benefits tended to be cautiously optimistic (European Commission 2008; Enderlein and Verdun 2009), the outbreak of the global financial crisis quickly required that assessment to be revised. The crisis severely challenged the legitimacy of EMU as it demonstrated that its benefits could not be taken for granted and that they might well involve conflicting interests between its member states. Notably, as the costs and risks of scaling back monetary integration were considered politically prohibitive, the primary crisis response was to create new European-level institutions to stabilize EMU. Thus, the initial institutional set-up was complemented by the mobilization of common resources in the form of bail-out programmes and by new forms of collective governance of economic and financial affairs (like the European Semester and Banking Union). In short, the crisis has made it clear that EMU is an institutional and political regime – something that it arguably already was from its inception – and taken away any suggestion that it can be considered as (merely) a rules-based system of macroeconomic governance that can be justified by its benefits alone.

However, if that is the case, there is an obvious need to reconsider the legitimacy basis on which EMU, and particularly its new institutional arrangements, operate (cf. Schmidt 2015). Even if these institutions have been established to ensure that EMU’s benefits are delivered, there is no going-back to the expectation that these benefits can be guaranteed or that they will be free from contestation among its members. Hence, we examine which sources of legitimacy the new institutional arrangements in EMU rely upon and, in particular, whether they provide for forms of input legitimacy to the extent that output legitimacy is not guaranteed or may run into political disagreement.

To this purpose, this contribution takes stock of the three main institutional innovations that have been introduced to stabilize the European Union’s economic and financial governance – namely the European Stability Mechanism, the European Semester and Banking Union (leaving the purely monetary operations by the European Central Bank aside – on those, see Chang 2020) – to identify the main similarities and differences between them and to come to a comprehensive assessment of the legitimacy basis of their operations. The actual analysis is preceded by a section that lays out our conceptual apparatus and the way we apply that to the three institutional arrangements. To provide a benchmark, a second section outlines schematically how, before the crisis, EMU could still be considered to be legitimised on the basis of outputs alone.

Our main finding is that the output-oriented approach that originally informed the legitimacy of EMU has remained prevalent after the crisis. In other words, the new institutions stay indebted to an understanding of EMU legitimacy that the crisis has revealed to be untenable. This is problematic as these new institutions come with new and potentially very intrusive powers that can be exercised over and among the member states. We thus argue for more robust channels of accountability and for increased input legitimacy both at the national and the European level.

2. Legitimacy in the European Union

It is no coincidence that the distinction between input and output legitimacy has gained particular popularity in the context of European integration. At the national level, the legitimacy of policies is usually not at issue given the democratic credentials of the national political systems. However, when competences come to be pooled at the European level, such democratic credentials cannot automatically be taken for granted. Even if the European Parliament (EP) was established as a directly elected supranational parliament, it was obvious that – for a long time – it not only fell short in actual powers but, more fundamentally, lacked the cultural and social infrastructure that could position it as the effective voice of the ‘European people’.

Fritz Scharpf (2006, 1) famously distinguishes between input and output legitimacy of ‘the exercise of governing authority’. Input legitimacy is the legitimacy that derives from the institutional arrangements – elections and consultation procedures – that ‘ensure that governing processes are generally responsive to the manifest preferences of the governed’ (idem). In contrast, output legitimacy derives from government for the people. Specifically, it reflects the extent to which citizens recognize policies as furthering their interests and as ‘effective solutions to common problems of the governed’ (idem).

Underlying the distinction between input and output legitimacy lurks the suggestion that the one can make up for shortfalls in the other. Still, one can assume legitimate policy-making to require at least a modicum of each. Democratic theory would further suggest that input legitimacy has some claim of normative primacy over output legitimacy, as it essentially responds to the fundamental value of individual autonomy; in the absence of input legitimacy, individuals remain subjects of the law rather than citizens that can see themselves as its co-author. In that respect, there is something in input legitimacy – a form of personal recognition – that no degree of output legitimacy can ever make up for.

However, as input legitimacy can already be deemed to be secured within member states through the usual national democratic arrangements, output legitimacy could take priority at the European level. Following this line, authors like Giandomenico Majone (1996) have insisted that European integration was to be restricted to regulatory domains where the adoption of common rules would be Pareto-efficient, that is, yield efficiency gains for all member states and their citizens. As long as European integration would remain within these domains, there was no reason to be concerned about a democratic deficit. Things change however once the European Union (EU) comes to adopt policies with redistributive consequences. These need some procedure of collective will-formation that serves to ensure that these decisions can also be justified to those who are negatively affected by them (Follesdal and Hix 2006).

As it became apparent that it was unlikely for EU policies to always be Pareto-efficient, there has been a mushrooming of ideas about alternative sources of legitimacy of EU decision-making (Héritier 1999). Output legitimacy remained of great importance, and much of the legitimacy of European integration as a whole continues to be staked on its presumed contribution to ensuring peace and economic growth. Institutional reforms have been carried through to provide EU decision-making with input legitimacy, ranging from the establishment and extension of power of the EP to increasing transparency and access to documents in EU decision-making. Beyond that, however, European institutions have explored alternative sources of legitimacy. One of these was the importance attached, particularly by the European Commission, to stakeholder consultation. Not only is stakeholder participation seen as an alternative form of citizen representation, it is also seen as bringing expertise to the table, which benefits the quality of policies and, by implication, their output legitimacy. In a similar vein, the Commission has had a tendency to pride itself on virtues like neutrality, independence, objectivity and expertise without the contamination of specific political interests, which are seen as further, technocratic, guarantees strengthening the EU’s output.¹ However, all of that can be read as essentially driven by the need to compensate for the inherent shortages in input legitimacy at the EU level; a characteristic that haunts EMU as well.

3. Democratic legitimacy of the pre-crisis EMU

From the perspective of legitimacy, a notable feature of EMU, as it started operation in 1999, was the marginal role assigned in economic policy to the EP, which, unlike the Council, enjoys the direct legitimation by European citizens (cf. Hodson and Maher 2002). Part of the explanation for this lack of a direct supranational channel for input legitimacy is that the output benefits of EMU seemed beyond dispute (Committee for the Study of EMU 1989; European Commission 2008). A single currency and the coordination of economic policies would bring efficiency gains in a more deeply integrated single market by reducing economic and financial transaction costs and facilitate upwards economic convergence among the member states involved. These benefits were logically expected to accrue to all members.

As far as input legitimacy was a concern in this context, much of it hinged on the choice of each state to join EMU; even if, beyond that willingness, admission to membership was conditional upon the state meeting certain criteria that were to ensure that it would not pose an economic risk for the other members. Thus, in line with the basic doctrine of ‘state voluntarism’ (Viehoff 2018), as they assented to the terms of the treaty, member states essentially entrusted democratic legitimacy on EMU policies. This is not merely a formalistic claim but normatively backed up by the democratic mandate that each government enjoys given the strong legitimation channels that guarantee citizens’ agreement at the national level.

In the context of EMU, the insistence of state voluntarism found its corollary in the principle of ‘debt liability’ (Viehoff 2018). This principle maintains that each state is individually liable for whatever debt it has contracted and for ensuring its ability to service that debt over time. Critically, this emphasis on national autonomy implies that other states take no responsibility for the economic consequences of national policy choices. The debt-liability view justified the belief that there was no need for a centrally managed ‘pot of money’ to be used in case of crisis. It furthermore translated into the adoption of a no-bailout clause (Article 125 TFEU), which vouched that in case a member state would find itself in financial trouble neither the Union nor any other member state would be available to offer solace.

Under these conditions, any suggestion that EMU policies required some sense of input legitimacy and collective will-formation at the European level was bypassed. However, reflecting the widely recognized asymmetry in EMU, this bypassing took a different form in the monetary and the economic pillar (see Verdun 1996; Hodson and Maher 2002; Howarth and Verdun 2020). As regards monetary policy it was assumed that the efficiency gains resulting from the attainment of price stability would outweigh the losses in terms of democratic legitimacy that come with the decision to hand the keys of the printing press to an independent supranational institution. The output legitimacy of monetary policy was moreover staked on the independence and credibility of the European Central Bank (ECB), which was difficult, if not impossible, to square with a conception of input legitimacy (Hodson and Maher 2002).

In contrast, in the economic pillar, it was impossible to deny the need for input legitimacy, but all of that was effectively pushed back from the European to the national level. Member states would retain the primacy over their economic policies at the discretion of the national democratic process. Still, to safeguard monetary stability and to avoid the imposition of costs by some member states on others, some overarching policy framework for the short and medium term was needed, including common ceilings for the maximum amount of government deficit and debt (cf. Committee for the Study of EMU 1989, 20). As long as these ceilings were observed, national economic policymaking was to be subject only to a soft process of collective policy surveillance. This set-up helped to ensure that the economic policy directions that were agreed at the European level would be depoliticised and primarily informed by technical judgments rather than political contestation. The Governing Council of the ECB decides on the level of interest rates, while the Commission assesses member states’ macro-economic policies according to the standards adopted. Thus, there was no need for European economic policy to become the object of a structured political bargain – like the one found in most national democracies, where the government proposes a budget and the parliament votes – since the agreed-upon rules were supposed to accommodate all ‘reasonable’ demands and to be apolitically applied by the Commission.

Thus, in many respects, EMU appeared as a kind of limited liability partnership rather than a union. Members committed to (a) share one and the same currency, (b) ensure their public finances would remain within the agreed bounds, and (c) retain individual responsibility for the economic and fiscal policies they would adopt within those bounds. In line with ‘state voluntarism’, this agreement derived its moral force from the input legitimacy expressed by member states’ initial consent.

Ultimately, however, the tenability of these agreements hinged on the persistent output performance of the Union. Once that performance could not be guaranteed, the basis of its legitimacy would fall through. In that sense, Hodson and Maher already correctly observed in 2002 that ‘The issue is to ensure that there is sufficient legitimation so as to prevent EMU being destabilized in the face of a crisis by the withdrawal of popular consent’ (400; cf. Verdun and Christiansen 2000). And that is exactly what happened.

4. The changed architecture of the post-crisis EMU

As the financial crisis hit the euro area in 2008 and 2009, its institutional framework was put to the test. Numerous banks became illiquid and even insolvent; governments felt forced to intervene in order to prevent contagion from one bank to the next; thus, public deficits went up, as did the spreads on government bonds. On top of that, the Greek government came to admit that it had falsified its financial data. As financial markets dumped Greek government loans, in early May 2010, the EU and the International Monetary Fund (IMF) stepped in to compose a package of loan guarantees (Pagoulatos 2020).

These bail-out arrangements evolved as time progressed, eventually resulting in the European Stability Mechanism (ESM). What is more, the EU started to work on institutions with a more preventive role so that EMU would not be as vulnerable to future financial crises. Most importantly from the viewpoint of collective political action, these involve the revision and extension of the Stability and Growth Pact which led to the establishment of the European Semester and the package of reforms that were to create a Banking Union. These are the three sets of institutions we focus on next.

4.1. Financial stabilization institutions

Because of the debt liability principle, the pre-crisis EMU did not foresee any form of reserve fund in case one or more of its members would lose access to credit. However, the financial crisis in Greece indicated that bankruptcy of an EMU member was an option and that would threaten the credibility of Euro as a whole. Under these crisis conditions, the decision was made that EU members and the IMF would provide loan guarantees to the government (Greece) in distress. Initially, these loan guarantees were agreed on a one-by-one, bilateral basis. By 2012, the ESM was established as the result of a process of bundling of functions and capacities of preceding financial stability facilities (the European Financial Stability Facility and the European Financial Stability Mechanism).

In the face of a direct crisis, the establishment of these financial stabilization institutions relied above all on output legitimacy. They were necessary to prevent the members of the euro area from experiencing a major decline in welfare; without them, EMU member states may well have gone bankrupt and the euro area disintegrated, which would have caused major economic disruptions (cf. Scharpf 2011, 183).

If the ESM can then be seen to serve an essential role in preventing the euro area from going over the financial precipice, whatever provision is made for input legitimacy is organized in an intergovernmental way. All (19) ESM members have a seat on its Board of Governors (usually the minister of Finance) and in its Board of Directors (usually a high-level civil servant), and all major decisions – above all the decision to disburse a loan – require unanimity. Thus, each European country has a political representative who can prevent that the country becomes committed to a decision for which there is no domestic support.

While this arrangement may look adequate given that it is mostly activated to prevent an immediate crisis, two important caveats need to be made. First, there are notable asymmetries among the ESM members in the extent to which their representatives are subject to parliamentary control and instructions. In about half of the EMU member states, the Minister of Finance can commit to the disbursement of loans or guarantees on her or his own accord while in the other half this decision requires an explicit mandate or ex post approval by the national parliament (Kreilinger 2015). These differences between member states partly reflect different national traditions in the way parliaments control the government. However, they are also indicative of power differentials between states, as is for instance illustrated by the fact that Germany as the greatest contributor also has granted the biggest role to its national parliament in approving any ESM disbursements (Howarth and Spendzharova 2019; Schild 2020). In that respect, there are significant – and, indeed, problematic – asymmetries in the democratic control that national parliaments can exercise over these decisions (Benz 2013).

Second, from the point of view of input legitimacy, it is unclear to which extent the democratic autonomy of a member state asking for financial support can be protected. Critical here is that such support is always conditional on a set of policy requirements in the form of so-called ‘macro-economic adjustment programmes’. These programmes involve extensive and detailed policy prescriptions that are considered necessary by the creditor countries (or, in practice, by the ‘troika’ – ‘the European Commission, in liaison with the ECB and, wherever possible, together with the IMF’ (Art.13.3 ESM) – to whom this responsibility is delegated) to remove the threat that the debtor poses for the rest of the Euro area. In line with the intergovernmental logic of EMU, the ESM treaty refers to this process as ‘negotiations’ rather than deliberation or collective will-formation, thus acknowledging that the two parties enter this process with contradictory interests: the debtor country wants to minimize the intrusion in its domestic policy choices while the ESM delegates want to remove any financial uncertainties. However, for the debtor country, the choice between accepting the macro-economic adjustment programme on offer or going bankrupt with all domestic consequences attached to it, is hardly a genuine and free one; especially if that offer comes in a situation of crisis that leaves little time for reflection.

Given the inherent power imbalance between creditors and debtors and the strongly intergovernmental nature of the ESM, there is a striking lack of institutional means to hold the policy-makers accountable. Obviously, whatever commitments are agreed are highly political, particularly as they can be expected to have significant disruptive and distributional effects in the economy of the debtor state involved. However, finance ministers sitting in the Board of Governors of the ESM are not accountable as a collective (Howarth and Spendzharova 2019) for the decisions they take on the funding and content of specific programmes. By extension, one wonders on what democratic authority the Commission (and the ECB and IMF) acts in these negotiations and to what extent the specific positions and priorities that it adopts can be said to ‘represent’ the collective interests of the peoples of the creditor countries, let alone that they take full account of the interests of the debtor countries (Notermans and Piattoni 2020; Pagoulatos 2020). In fact, the way the finance ministers delegate this responsibility and the way the task is phrased in purely functional terms ignores the political implications and appears as a way for the ministers to distance themselves from any political responsibility for the outcome of the negotiations. What is more, the EP has been completely left out of the ESM’s architecture; its involvement is limited to being part of an ‘economic dialogue’ with representatives of the ECB, the Commission and the IMF (Regulation 472/2013).

Even though all countries have by now left the bailout programmes, the unpredictability of financial shocks leaves the possibility that sooner or later resort will need to be made to the ESM. While there are few alternatives from the perspective of output legitimacy in such a situation, there remain major issues of input legitimacy; both as regards the position of the debtor country but also in the positions that are adopted by the creditor countries. While the euro area cannot control the ebbs and flows of global finance, it can increase its internal preventive capacities. That is where we turn next.

4.2. The European Semester

Beyond the bailouts of individual governments, the most structural EU policy response to the euro area crisis has been the stepping up of macro-economic surveillance of its members through a number of EU legislative packages (the so-called ‘Six-Pack’ and ‘Two-Pack’) culminating in the ‘European Semester’ (Verdun and Zeitlin 2018). Substantially, with the addition of a ‘Macroeconomic Imbalance Procedure’ (MIP), the scope of European surveillance has been broadened beyond public finance to also include macro-economic indicators (inter alia, private debt, liabilities in the financial sector, and unemployment). Furthermore, most notably, the European Semester harmonizes the different review procedures in a single coordinated calendar around several clearly marked milestones.

As such, the European Semester builds upon the multilateral surveillance of economic policies as it already took place before the crisis, but it has reinforced its preventive role in identifying any parameters in member states’ finances and economic order that may make them vulnerable and threaten the stability of the euro area as a whole (Haas et al. 2020). In that sense, there is in principle a double claim to output legitimacy here, as citizens in the EU are not only served by the stability of the euro area but also by the policy recommendations that come out of the process and that may help increase the quality of member state government policies.

Still, these likely increases in output legitimacy come at a price in terms of input legitimacy. Crucially, while much of the European Semester can be characterized as ‘soft policy’, ultimately there remains the threat of penalties. In fact, in the wake of the crisis, the sanction potential of the procedure has been increased by reversing the decision rule to the effect that Commission recommendations are adopted unless this is prevented by a qualified majority of the EMU member states.² While no financial sanctions have been imposed so far, the more direct threat that countries face is the prospect of more intensive policy scrutiny and ever more detailed prescriptions. Thus, Iain Begg (2013) has aptly characterized the new EU economic governance as ‘a system that relies primarily on deterrence’.

By conditioning national economic plans very early in the process, the European Semester acts as a constraint, not only on governments’ fiscal policy (Dawson 2015; Scharpf 2011), but also on the power of national parliaments to contest government decisions. Under these conditions, by the time it gets to the actual budget in the autumn, national parliamentarians are unlikely to wield their veto. Notably, however, also already in the earlier stages of the process, national parliaments are generally content to be consulted on the pre-commitments of their governments (in the form of the Stability and Convergence Programmes and the National Reform Programme) and do not insist on the power to amend and veto these. In all, the European Semester puts national parliaments at the losing end of a considerably ‘reinforced two-level game’ (Crum 2018 drawing on Putnam 1988).

If input legitimacy is thus lost at the national level, it is not compensated for at the EU level. The powers of the EP in the European Semester are consultative and advisory only. However, since most of the substantive direction of the process and all decisions on sanctions and tightening surveillance are taken at the European level, there is a clear question as to how these decisions can be legitimated. From the viewpoint of parliamentary accountability, the main concern here is that the structure of the European Semester makes it very difficult to determine the locus of political authority at the supranational level (cf. Crum and Curtin 2015, 82f.): while the governments, who are formally responsible, cannot be held accountable as a collective, the Commission, that calls most shots in practice, operates under the pretence that these are merely administrative decisions. As a consequence, no one is accountable (Crum 2018; Hodson 2020).

Much of this accountability gap is glossed over by the reliance on quasi-objectified rules and benchmarks (Schmidt 2015), like the famous 3 per cent public deficit constraint imposed by the Stability and Growth Pact (SGP). However, this rules-based understanding of the exercise of authority in the European Semester is difficult to sustain in practice (Schmidt 2016). The rules that have been adopted and the levels at which benchmarks have been defined clearly have political implications. As a self-proclaimed ‘highly political’ college (Juncker 2014), the European Commission led by Jean-Claude Juncker admitted that the application of these rules cannot be done in a purely objective way and inevitably requires some flexibility and context-sensitive judgment (European Commission 2015). Still, it quickly retreated from any suggestion that it could act on its own political authority as it underlined that the credibility of the SGP requires it to operate as a ‘rules-based system’ (ibid, 4).

The recognition of the indispensability of context-sensitive judgments exposes a gap in supranational political authorization, as it means that it is impossible for the Commission to be led exclusively by the instructions from the Council (cf. Dawson 2015; Schmidt 2015, 99). In practice, the two institutions aim to limit the ‘disjunction between discourse and action’ in the administration of the policy (Schmidt 2016, 1033). The Commission treads cautiously in issuing its recommendations and seeks to anticipate as much as possible the will of the (majority of the) member states. In turn, the member states have been quite happy to leave the lead to the Commission and to adopt its proposals as a rallying point. Typically, even when it comes to the adoption of the political priorities at the March European Council, the Heads of State or Government tend simply to endorse the priorities of the Commission’s Annual Growth Survey rather than to formulate their own (e.g. European Council 2016, 3).

Undeniably, the European Semester involves the making of political judgments at the European level, which effectively constrain member states’ financial and economic decisions without enjoying the necessary democratic legitimation. Decisions are delegated to the European Commission, which operates on the basis of political rules that require the exercise of considerable discretion in their application. Even if the member governments in the Council were to put their full political authority behind these judgments, they would only be accountable individually and not as a collective.

4.3. Banking Union: SSM and SRM

While the great recession of 2008 morphed into a highly salient sovereign debt crisis in the euro area, the real causes of financial instability were to be found in the banking sector. Initially, the decision to share the same currency and to create a supranational central bank did not have major consequences for the institutional setting of banking regulation and supervision. National authorities enjoyed much autonomy in carrying out their tasks and could follow their own ‘philosophy’ of supervision, creating considerable room for regulatory arbitrage by credit institutions. At the same time, a vigorous banking sector encouraged national authorities to be lenient towards their own banks, so as not to penalize their local ‘champions’ (Véron 2015). Together, national discretion and ‘banking nationalism’ created the incentive for excessive risk-taking by banks. Thus, the EU accommodated a highly dynamic and interdependent financial sector without any substantial coordination of regulation.

This mix became problematic as soon as the financial crisis reached the EU. Member states decided to bail out their credit institutions in an uncoordinated fashion, forcing their neighbours to do the same (Pisani-Ferry and Sapir 2010). Moreover, the cost of these bailouts cast doubts on the solvency of the member states themselves, thus shifting the problem from the banking sector to the government bond market. However, lower bond prices further weakened bank balance-sheets and their capacity to lend.

The negative feedback loop between banks and sovereigns quickly took the centre stage in the euro area and paved the way for initiatives to establish a Banking Union (Véron 2015; Howarth and Quaglia 2016, 2020). On the one hand, the problem of cross-jurisdiction externalities was solved by giving the ECB the task of supervising all major credit institutions using Article 127(6) of the TFEU as legal basis. Under the Single Supervisory Mechanism (SSM) regulation, the ECB acquires the power to grant banking licenses and to supervise credit institutions. A new body, the Supervisory Board, has been created inside the central bank to carry out most supervisory tasks.

On the other hand, the problem of the sovereign-bank nexus was tackled by a resolution framework that would make shareholders bear the cost of their institutions’ losses, thus avoiding the costly bailouts that pushed member states to the brink of insolvency. This ‘second pillar’ is constituted by the Single Resolution Mechanism (SRM). Decision-making on the resolution of the euro area’s largest banks has been centralised within the Single Resolution Board (SRB), a committee of six independent members appointed by the EP on the basis of a proposal by the Commission. This new institution decides on resolution actions and, eventually, on any use of the Single Resolution Fund, which is currently being build up and may be employed to ease any resolutions and to compensate exceptional losses of shareholders and creditors.

In the design of Banking Union, we recognize the contractarian inclination to minimize the political nature of both the responsibilities established and of the regulations to be applied.This justifies the central role assigned to apolitical, expert-based institutions. Just as the European Semester, Banking Union has been organized with the aim to minimize externalities and moral hazard through ex-ante regulation, in order to avoid imposing financial burdens on other member states. At the same time, instruments like a European Deposit Insurance or a European fiscal stabilization instrument, which inevitably would have highlighted the political nature of the issues involved, have been shunned (Howarth and Quaglia 2020).

Furthermore, the regulatory responsibilities have been organized away from the European Commission and assigned to the ECB and the newly established SRB instead. Obviously, however, the political salience of banking failures and the intrusiveness of the policy tools employed by the supervisory authority make banking supervision very different from the original tasks of the ECB in the management of interest rates and monetary aggregates. Notably, according to the SSM Regulation, the ECB will have investigatory powers similar to those granted in the area of competition to the Commission (Wolfers and Voland 2014); it will be able to impose fees, carry out on-site inspections and impose sanctions on credit institutions. The impact of such decisions can be significant, as is illustrated by the somewhat awkward manoeuvring of the ECB around the cases of the Banca Popolare di Vicenza and Veneto Banca in 2016, which were eventually left to be wound up under Italian law, as well as by its more resolute resolution of the Spanish Banco Popular in 2017.

All of this would suggest that banking supervision should be subject to a higher degree of parliamentary scrutiny than the original monetary responsibilities of the ECB. Nevertheless, the assignment of new powers to the ECB has failed to translate into a substantial revision of the scrutiny powers of the EP and of the Council. Article 20(1) of the SSM Regulation states that ‘the ECB shall be accountable to the European Parliament and the Council for the implementation of this regulation’. However, the precise substance of the term ‘accountability’ remains under-specified (Amtenbrink and Markakis 2019). The text of the regulation suggests that accountability in both supervision and monetary policy is mostly an exercise of transparency and dialogue (e.g. Fraccaroli, Giovannini, and Jamet 2018). Presumably, the ECB is expected to explain what it does and how it does it, but there is no way the EP can remove executives from office or revise the mandate of the institution. We should note, however, that despite the EP’s lack of ‘hard’ powers over the ECB, the formal channels for information sharing between institutions have gained in importance (see Fraccaroli, Giovannini, and Jamet 2018; Amtenbrink and Markakis 2019; Ter Kuile, Wissink, and Bovenschen 2015). In addition, every year the ECB publishes a report on its supervisory activities, which is presented in the EP and sent to the Eurogroup, the Council, the Commission and national parliaments.

Thus, Banking Union has been established as a regulatory capacity at great distance from political guidance. As its main goal is to reduce systemic risks in the financial sector through the supervision of the largest banks in the euro area and as it operates complimentary to national oversight institutions, it is clear that it relies above all on a claim to output legitimacy at the supranational level. While Banking Union in this form certainly contributes to strengthening the sector, it remains to be seen whether its competences are sufficient if another financial crisis would hit. Furthermore, by choosing to lodge significant new powers with the ECB, these are subject to an extremely light system of accountability that was designed for monetary policy rather than for supervisory tasks.

5. Conclusion: pathways to post-crisis input legitimacy in EMU

This contribution set out to take stock of the main institutional innovations in EMU after the crisis and to assess to what extent their operation can continue to rely on output legitimacy as the pre-crisis EMU did. The financial crisis has made it clear that EMU has come with a much higher level of interdependence than the existing institutional structure was prepared for. It also shows that this interdependence does not necessarily result in beneficial effects for all but can involve negative financial spill-overs and externalities (cf. Schmidt 2015). Reflecting this, the battery of institutions that have been created over the course of EMU’s second decade has fundamentally enlarged the scope of its policy domain (see also Hodson 2020).

When we examine the sources of legitimacy on which these new institutional arrangements presumably rely, we find a strong emphasis on output legitimacy, very much in line with the legitimacy conception that shaped the institutional setting of the pre-crisis EMU. While new competences were established at the European level, much effort has been made to depoliticize them, inscribe them in a rule-based framework and have them handled by technocratic bodies. The administration of Banking Union is left to the independent bodies of the ECB and the SRB. Technocratic institutions design the details of ESM lending, while the Commission is assigned the role of applying the new set of rules to member states’ economic policies as part of the European Semester.

Depoliticization and the expansion of executive capabilities do not necessarily pose a problem of legitimacy. However, with this expansion of supranational competences, there are unmistakably political questions seeping through which require justification and protest against the delegation to technocratic bodies. Admittedly, there are variations in the extent to which these concerns arise. They appear, for instance, more prominently in the European Semester than in the context of Banking Union. The case of the ESM is more exceptional, but obviously, there is a major political asymmetry at its heart in the opposition between debtor and creditor countries.

In none of the three cases discussed have systematic institutional arrangements been set up to ensure that such political decisions are subject to adequate channels of input legitimacy. In the ESM, democratic input is essentially marginalized except for the few major creditor states. In the European Semester, we find the ability of national parliaments to control the budgetary process constrained without the creation of any kind of political arena at the supranational level that can compensate for that loss. In Banking Union, regulatory power has been uploaded to independent European agencies without much consideration of the political implications that these powers have. In general, we find that the power exercised by the new institutions cannot be meaningfully held to account by those subject to it. The ESM Board of Governors is not accountable for its decisions as a collective; political responsibility seems to be lost in the complex process of the European Semester; and there are limited institutional channels to punish the ECB if it were to abuse its new powers as bank supervisor. The SRB is somewhat of an exception in this regard as, notwithstanding its strictly delineated task, it can be monitored rather closely by the EP.

Nevertheless, the overall picture that emerges is that the traditional fora for democratic representation have mostly lost power in the post-crisis era. One obvious way to reverse this trend is to turn the soft powers that the EP has now been given in ‘economic dialogues’ into more decisive powers to veto and amend decisions of the ESM, the Commission, the ECB and the SRB. Such parliamentary controls at the European level do not necessarily involve the communitarisation of the policies. They can co-exist with continued controls by the member states in the ESM and the Eurogroup. At the same time, there is considerable scope to increase the democratic scrutiny of the national governments by increasing the transparency of the proceedings and by improving the strategic timing and focus of the engagement of national parliaments. Thus, much like the EU legislative process, also EMU policies can in principle be subject to two channels of parliamentary control that operate as a double democratic lock on the policies that are adopted (cf. Crum 2018; Hennette et al. 2019).

In conclusion, it seems that the new EMU architecture will be more intrusive yet more distant from European citizens as we witness a mismatch between the powers exercised by the new institutions and their legitimation. For sure, the crisis required EMU to be reinforced by new institutions. However, while the competences and powers of EMU institutions increased considerably, the legitimation channels have broadly remained the same. The conceptual problem lies in the apparent conviction that the rules’ supposed output legitimacy will be enough. This conviction seems however impossible to uphold after the experience of the crisis. Sovereign bailouts, the resolution of credit institutions and the constraints on fiscal policy are all areas with highly salient distributive implications and little room for Pareto-optimal solutions. In these situations, technical approaches driven by rules and agencies are not enough to ensure optimal policy responses, simply because the recognition of a policy’s optimality has to be preceded by a moment of collective will-formation in which parties justify to each other which criteria should be used to evaluate the policy’s outcomes. The appropriate fora for this democratic exercise are not intergovernmental institutions or technocratic bodies that lack clear lines of accountability, but parliaments.

Acknowledgments

An earlier version of this paper was presented at the EMU at Twenty workshop at Leiden University (16-17 November 2018) https://www.uvic.ca/interdisciplinary/europe/eu-grants/network/eurosem-18-21/the-emu-at-twenty/index.php. This workshop was supported by the Institute of Political Science at Leiden University and launched the Jean Monnet Network entitled “The Politics of the European Semester: EU Coordination and Domestic Political Institutions (EUROSEM)”, Agreement number: 600110-EPP-1-2018-1-CA-EPPJMO-NETWORK (Grant agreement nr 2018-1359), cofounded by the Erasmus+ programme of the European Union. The paper has benefitted from the enthusiastic feedback of the workshop participants and, in particular, from comments from our discussant Reinout van der Veer, the editors of this volume, Amy Verdun and David Howarth, and two anonymous reviewers. Ben Crum’s research for this paper has been carried out as part of the RECONNECT project, which has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 770142.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This work was supported by the H2020 Societal Challenges [Grant Agreement no. 770142].

Notes

1. Vivien Schmidt (2013) has advocated the use of a separate category of ‘throughput legitimacy’ to capture these procedural virtues. For the sake of simplicity, we prefer to stick to the distinction between input and output legitimacy, where we consider the mentioned values as subservient to the quality of the EU’s output.

2. The one exception in this regard is the launch of an excessive imbalance procedure under the MIP, which requires the usual qualified majority in the Council.