Keywords

JEL Classifications

European Banking Union

European banking union (henceforth banking union or BU) introduces for the first time an integrated approach in supervision and resolution of European banks, representing an important step towards enhancing economic and monetary union. The aim of banking union is to deliver absolute consistency of implementation of new regulatory rules across the euro area (at the time of writing 129 banking groups, representing more than 80% of the euro area banking sector’s assets; ECB (2015)), ensuring that the financial institutions of all member states will be subject to a single supervision, a single resolution and a common deposit insurance system. The need for a banking union emerged in response to the 2007/08 global financial crisis and the ensuing sovereign debt crisis in the eurozone. In particular, the sequence of events highlighted the costs of the vicious link between public and private sector debt, and how these can easily overflow national borders and cause systemic risk and failures.

The banking union proposal dates back to June 2012; it covers a preventive stage (regulation and supervision), and a crisis management stage (resolution and safety nets) (European Commission 2012; IMF 2013b).

The first two components of BU, a single European supervisor (the Single Supervisory Mechanism; SSM) and a single resolution authority (the Single Resolution Mechanism; SRM) have been agreed. The third element of BU, however – a European deposit insurance scheme covering eligible individual deposits in all participating countries – has been stalling, largely because of political opposition from some creditor member states (Germany in particular).

The SSM has been in place since 4 November 2014; this is the date from which the European Central Bank assumed responsibility for bank supervision. The SSM is a key ingredient of the BU, but it is not the only one. In particular, a European approach to the resolution of banks – with the SRM centred on the idea of a Single Resolution Board (SRB) and a Single Resolution Fund (SRF) – needed to follow. The EU adopted a Bank Recovery and Resolution Directive (BRRD) together with an agreement on the SRM from December 2013. This agreement – following on from the SSM already agreed – was significant because it meant that two of the three components of BU have been operational since 2015. Nonetheless, both elements of BU have been somewhat watered down from their original conception. The SSM will de jure not be supervising the whole European banking system, with national authorities continuing to supervise smaller financial institutions. Furthermore, unlike the SSM, the SRM will be ‘single in name only’ (Posen and Véron 2014) as the framework that sets up the resolution mechanism foresees a significant degree of continuing autonomy for national authorities (see the section below on ‘Progress Towards Achieving a Banking Union’) – particularly on the issue of funding – at least for the next eight years. Progress has been very uneven on the third element of BU as well, with a common approach to deposit insurance having been sidelined during the first stages of the negotiations. Despite a first legislative proposal for a euro-area wide protection for bank deposits that came as late as 24 November 2015, negotiations are currently stalling. The lack of this third element is critical because it means there will ultimately be no European backstop for depositors in the event of a new banking crisis.

If implemented properly, the original vision for BU may be the most far-reaching reform since the inception of the euro (Constâncio 2013). The fact that the BU vision was further developed in the European Commission’s blueprint for economic and monetary union (Juncker et al. 2015) reveals the Eurozone’s willingness to continue to deepen integration and to put in place a framework making member states’ participation in the eurozone ‘sustainable’ (see also Pisani-Ferry 2012). However, with an established supervisory authority, a resolution mechanism on the way and a delayed agreement on a common deposit insurance scheme, it remains to be asked whether the European banking project can be credible without a fully fledged fiscal backstop.

Background to Financial Supervision in Europe

Financial market regulation under the Basel Accords, as well as the system of EU financial supervision before 2010, were generally characterised by the lack of mutual recognition. The existing Lamfalussy Process envisaged a largely delegated legislation and enforcement system with an explicit legislation in co-decision procedures (see also ECB 2010). The implementation and transposition of detailed rules on supervision and resolution were delegated to three Committees – the so-called 3 Level Lamfalussy (3 L3) Committees: the CESR (Committee of European Securities Regulators), the CEBS (Committee of European Banking Supervisors) and the CEIOPS (Committee of European Insurance and Occupational Pensions Supervisors). Day-to-day supervision was left to national supervisory agencies, with a strict separation of supervision from central banking, both geographically and functionally (see also Masciandaro et al. 2013).

After 2010, such an approach to financial supervision and regulation changed, under the pressure of the systemic nature of the crisis and the de Larosière report. On the legal side, there was a significant tightening of the regulation of banks, with Basel III raising minimum capital ratios and redefining riskiness of assets. Furthermore, the de Larosière Report (de Larosière Group 2009) established a European Systemic Risk Board (ESRB), chaired by the ECB’s President and Vice-President, with the aim of providing macroeconomic supervision of the financial system as a whole. (For a discussion of the governance of the ESRB see Gerba and Macchiarelli (2015).) The ESRB was created at the end of 2010 as a part of a new two-pillar system of financial supervision, the European System of Financial Supervision (ESFS). The report also gave recognition to three European Supervisory Authorities (ESAs) to cover micro-prudential supervision, representing the ESFS second pillar. These three EU-level bodies, being effective as of 1 January 2011, were not created ex novo, but they upgraded the existing 3 L3 Committees. In particular,

  • the CEBS was upgraded into the European Banking Authority (EBA);

  • the CEIOPS was upgraded into the European Insurance and Occupational Pensions Authority (EIOPA); and finally

  • the CESR was upgraded into the European Securities and Markets Authority (ESMA).

This change in governance structure marked not only the beginning of a greater (than in the rest of the world) involvement of the central bank in Europe, but also the start of a two-pillar strategy ensuring – by means of institutional separation and coordination with national supervisors – a system of checks and balances between macro- and micro-prudential supervision (see also Goodhart and Schoenmaker 1995; Masciandaro et al. 2013; Eijffinger 2013; Goodhart 2014).

The first agreement on banking union came in September 2012. The European Parliament’s final ‘go-ahead’ for the ECB to be fully entrusted with responsibility for the supervision of banks in the framework of the SSM came after extensive negotiations between various stakeholders. This happened one year after the first agreement, on 12 September 2013. The 2012 EU Council agreement appropriately conferred broad investigatory and supervisory powers on the ECB, which – as of November 2014 – is responsible for the effective and consistent functioning of the SSM. National authorities remain responsible for the banks remaining under their direct supervision (the so-called ‘less significant financial institutions’).

Guidance on the design of an effective supervisory mechanism for Europe was provided in the Basel Core Principles (the so-called ‘Core Principles for Effective Banking Supervision’; http://www.bis.org/publ/bcbs30a.htm). According to these principles, a number of preconditions and prerequisites were to be met at the euro area level, including (1) the implementation of coherent and sustainable macroeconomic policies; (2) a clear framework for financial stability policy; (3) an effective crisis management and resolution framework to deal with bank failures and minimise disruptions; (4) an adequate safety net to deal with confidence crisis while minimising distortions; (5) a well-developed public infrastructure; and (6) effective market discipline. On the other hand, as underlined by IMF (2013b), prerequisites to establish a sound basis for the SSM included: (1) its operational independence; (2) clear objectives and mandates; (3) legal protection of supervisors; (4) transparent processes, sound governance and adequate resources; and (5) accountability (see also IMF 2013b; Gerba and Macchiarelli 2015). As we shall discuss in the following sections, after the comprehensive assessment performed by the SSM at the end of 2013, with extensive granular balance-sheet facts being provided and a higher degree of transparency and availability of information to the public, the SSM seems to meet these criteria.

The European ‘Doom Loop’

The crisis highlighted the importance of having in place a framework for dealing efficiently and in a timely manner with the resolution of cross-border financial entities (Obstfeld 2013), avoiding the long-term implications on fiscal sustainability of having national governments and banks dangerously tied together (see, inter alia, Reinhardt and Rogoff 2013; Gennaioli et al. 2014). These ties essentially intensified during the eurozone’s crisis for two reasons. First, banks engaged in carry-trade by using ‘cheap’ central bank liquidity to purchase government bonds (Acharya and Steffen 2015). (Central bank liquidity came mainly in the form of three-year long-term refinancing operations (LTRO), with the interest rate fixed at the average rate of the main refinancing operations at the time (1% p.a.) and full allotment of the bids (for further technical details see ECB 2011).) Second, there was a rapid rebalancing of banks’ international portfolios towards ‘home’ assets and bonds (Battistini et al. 2014; Valiante 2015). The latter was possibly the result of risk-shifting (Gennaioli et al. 2014; Farhi and Tirole, 2016; Acharya et al. 2015); discrimination (Broner et al. 2016); and financial repression (Chari et al. 2014; for a general discussion see also Reinhart et al. 2011).

Government guarantees to banks at the expense of higher debt and the inability of regulators to stall the crisis, together with a ‘faulty’ design of the currency union – centred on a single central bank and multiple Treasuries (see De Grauwe 2016) – are known to be amongst the weighty factors at the root of the private-public European ‘doom loop’.

In the euro area, in particular, together with the impossibility of monetising debt (an explicit provision contained in the Lisbon Treaty on the Functioning of the EU – the ‘no bailout rule’ – art. 125), as of 2010 countries had de facto to compete ‘internally’ over capital flows (Valiante 2015). This was the reflection of an institutional setup built on the idea of ‘tying one’s hands’ – i.e. guarding against government failure by simply agreeing on strict fiscal rules (e.g. the Stability and Growth Pact) and letting markets find their equilibria (Fuest and Peichl 2012). (A key reason for the failure of international capital markets to differentiate sufficiently between countries according to the state of their public finances was that the ‘no bailout rule’ was just not credible (Fuest and Peichl 2012). In other words, before 2010 financial markets simply did not set incentives to limit government debt in the Eurozone, and very small borrowing premia were to be paid over German safe-haven rates.) The sovereign debt crisis that followed confronted almost all non-AAA-rated euro area countries (Greece and Ireland first, followed by others by 2012) with a liquidity dry out, as the result of a flight-to-quality of capital – facilitated indeed by the single currency – towards their ‘safer’ EMU peers. This translated into higher public borrowing costs, a frailer banking system and overall larger bailout charges ex post.

Figure 1 proposes a stylised representation of the aforementioned European doom-loop. This representation does not consider contagion or spillover effects from, or to, other countries, being broadly related to recent literature on doom loops in closed economies (see e.g. Acharya et al. 2015). In this representation, whatever the entry point is (private sector leverage, unsustainability of public finances, lack of structural reforms) there is a self-reinforcing effect relating to the classical problem of (ir)rational runs in which the market can push an economy into a ‘bad’ equilibrium (see also De Grauwe and Ji 2013). This amplification within the EMU had to do, firstly, with a collapse of confidence in certain markets and institutions at the same time, and the broader fragility of financial systems, because of increased counterparty risk or asymmetry of information (see also IMF 2013a). Secondly, it was linked to the distressed financial sector inducing government bailouts (or private sector deleveraging; see Acharya et al. 2015). The latter, in particular, created a vicious interaction between asset prices (via banks’ balance sheets) and borrowing constraints (Borio and Zhu 2012; Brunnermeier and Pedersen 2009; De Grauwe and Macchiarelli 2015), where – simplifying – the fire sale of government bonds in some countries (as the result of confidence loss and excessive debt taking) increased sovereign credit risk, in turn weakening the financial sector, with an ensuing liquidity dry out and freezing of lending to the real economy. Overall this eroded bond holdings and the value of government guarantees, requiring further support, and so on.

European Banking Union, Fig. 1
figure 565figure 565

A stylised representation of the European ‘doom-loop’

Full size image

Why a Banking Union for Europe?

The governments’ last-resort guarantees to their own financial institutions were initially granted in an uncoordinated manner within the EU. Government asset support mainly took two forms (see also Gros and Schoenmaker 2014): asset insurance schemes, which maintained the assets in the banks’ balance sheet, and asset removal schemes, which transferred the assets to a separate institution (bad bank). Purchases of impaired assets often occurred after earlier government capital injections. In the case of bank debt guarantees, approximately half of those that received capital injections also received government guarantees for their bank debts. Coordinated support happened only later and was led by the European Commission in the context of its State Aid policy, with the aim of preserving an EU integrated financial market. Before the Commission launched a bank recovery proposal, a number of EU countries, including Austria, Belgium, Denmark, France, Germany, Ireland and the UK had already put in place new rules for the resolution of their distressed banks. Such repeated bailouts not only increased sovereign debt, but also imposed a large encumbrance on taxpayers. The state aid measures that were used, in the form of recapitalisation and asset relief measures between October 2008 and December 2012, amounted to €591.9 billion or 4.6% of EU 2012 GDP, with the highest share belonging (in order) to Ireland, the UK and Germany (European Commission State Aid Scoreboard’s (2013) figures. Available at http://ec.europa.eu/competition/state_aid/scoreboard/amounts_used_2008-2012.xls). Including approved aids and guarantees, this figure jumps to over 12% of EU GDP for the period 2008–12 only. In the euro area, 37% of capital injections and 63% of the asset relief measures were granted to the three largest financial institutions (see also Gros and Schoenmaker 2014).

Beyond government upkeep, central banks provided unprecedented liquidity support to illiquid (and insolvent) banks as well. Specifically, the European Central Bank during the first stage of the crisis focused its programme – albeit not exclusively – on direct lending to banks (see the preceding section), reflecting the bank-centric structure of the euro area financial systems (see also Gabor 2014; ECB 2014). This was different from the Federal Reserve and the Bank of England, which expanded their respective monetary bases largely by purchasing bonds in the first place.

Looking at the recent history of bailouts, the advantage of a permanent bank supervision and resolution framework, as compared to the ad hoc measures that were employed during the crisis, primarily resides in its transparency regarding the list of eligible institutions and the conditions of access to funding. Second, it introduces a limitation to free-riding derived from unlimited recourse to public money, allowing overall a balanced burden-share between private investors and taxpayers, possibly resulting in lower funding costs ex post. At the same time, the BU proposal recognises the systemic nature of risk facilitated by the single currency, and the potential dangers and domino effect that ‘systemically important’ financial institutions would have, given their cross-border reach, within the E(M)U (see also Obstfeld 2013; Gros and Schoenmaker 2014; Goodhart 2014). Finally, the proposal acknowledges the issue of the moral hazard of national governments both over time – with a tendency to offload the costs of restructuring the domestic banking sector to future governments – and across countries – particularly, relying on the ECB’s and European Stability Mechanism’s last resort support. The latter two points relate to the literature analysing the combination of limited commitment on the part of the government ex ante, and the possibility of bailouts ex post (see, among others, Acharya and Yorulmazer 2007; Chari and Kehoe 2016; Farhi and Tirole 2012). In particular, this literature highlights a mechanism by which government bailouts are provided only when a sufficient number of financial institutions are in trouble ex post, so that strategic complementarities in financial risk-taking arise: i.e. individual financial institutions may engage in higher financial risk-taking ex ante the higher the collective risk-taking, as this increases the likelihood of a government bailout ex post. The existence of such complementarities and systemic risk thus provides a rationale for macro- and European measures. (Broner et al. (2016) put forward another rationale for a BU: a BU is thought to reduce discrimination between domestic and foreign investors.)

Legal Underpinning

The legal foundation of BU is contained in a single rule book made up of three main elements.

  1. 1.

    A set of rules on capital requirements (Capital Requirements Directive – CRD IV).

These rules entered into force on 1 January 2014, and replaced the original Capital Requirements Directives (2006/48 and 2006/49), transposed the international Basel III agreement into EU regulation and ensured that banks hold a sufficient buffer to withstand potential losses.

  1. 2.

    The proposal for strengthening the Deposit Guarantee Schemes Directive (DGSD) (Revision of Directive 94/19/EC).

The aim of the latter was to harmonise and simplify deposit guarantee rules in the EU and improve the functioning of the existing guarantees across the board, with protection of deposits up to €100,000 (from the existing €20,000 limit). According to the directive, all credit institutions will be required to join the DGS instituted at the national level. The Council has reached a political agreement with the European Parliament on the revised directive, with the Parliament formally adopting this revision in April 2014.

  1. 3.

    Bank Recovery and Resolution Directive (BRRD) (Directive 2014/59/EU).

This directive gives powers to authorities across the EU to act effectively to prevent bank crises and to ensure orderly restructuring and resolution in the event of bank failure. The aim is to avoid negative effects on financial stability and to reduce recourse to taxpayers’ money, avoiding replicating the scenario seen during the first stage of the crisis. The directive followed a Commission proposal in June 2012. The European Parliament and the member states reached an agreement on 11 December 2013. These new rules, which entered into force on 1 January 2015, established that the costs of bank failure will in the first instance be borne by bank shareholders and creditors, according to a clearly defined hierarchy, and thereafter met from dedicated resolution funds held by each member state.

Progress Towards Achieving A Banking Union

Common Bank Supervision

A European single supervisor (SSM) became operational in November 2014 (see section on ‘Background to Financial Supervision in Europe’). Under the SSM, responsibility for bank supervision in the euro area was shifted from national authorities to the European Central Bank. The ECB is in charge of supervising ‘systemically important’ banks directly (equal to more than 80% of euro-area banking assets, including banks with over €30 billion in assets or 20% of national GDP, or ‘if otherwise deemed systemic’). National authorities will continue to supervise smaller financial institutions. (In September 2014, the ECB published the list of significant supervised entities. The latest release (31 May 2016) with change in significance for some banks is published here: https://www.bankingsupervision.europa.eu/ecb/pub/pdf/list_of_supervised_entities_20160331.en.pdf.) The latter arrangement was essentially a political one, championed by some member states – Germany primarily – wanting to keep direct monitoring of ‘local’ institutions. The federal approach that emerged as a concession to local banks’ lobbies highlights how banks’ management in some countries cultured a strict affiliation with the political establishment and local electorate (Valiante 2015), largely through ‘not-for-profit’ credit institutions such as foundations (e.g. the Spanish Cajas and the German Landesbanken). Overall, however, while smaller banks were de jure exempted from direct SSM supervision, the €30 billion threshold has de facto left the majority of the eurozone banking assets under the SSM’s umbrella – including almost all German Landesbanken (Posen and Véron 2014). Furthermore, the ECB will set and monitor supervisory standards and work closely with the national competent authorities for these banks, with the option of expanding its remit and supervising them directly in order to ensure that SSM standards are applied consistently (ECB 2014, 2015).

To conclude, while the design of a common bank supervisor is far from faultless, given the challenge to financial stability that small financial institutions may pose, these challenges in terms of supervision are, for the moment, not large. (It is worth noting that the majority of ‘local’ banks in the EMU are concentrated in Germany, and, to a lesser extent, Austria and Italy; see Véron’s blog entry on Bruegel: ‘Europe’s Single Supervisory Mechanism: Most small banks are German (and Austrian and Italian)’, 22 September 2014.) The current SSM design represents an adequate compromise given the existing trade-off between political feasibility and economic ‘first-best’ in Europe. In addition, achieving a truly single market in banking services will possibly require more time than a couple of years, with further supervisory initiatives, as well as regulatory and legislative steps, having to be adopted in the future (see also Schoenmaker and Véron 2016).

The Single Resolution Mechanism

The SRM was first proposed by the European Commission in July 2013. This mechanism came to complement the SSM as of 1 January 2015. Countries joining the SSM are to join the SRM too, which means that the SRM applies to banks in the euro area member states under the SSM, plus those EU countries wishing to opt in. The SRM is built on the national resolution authorities established by the BRRD. The SRM aims to ensure that if – despite SSM supervision – a bank faces serious difficulties, its resolution would be managed in a centralised manner, with minimal cost to taxpayers and the real economy; which is one of the focal points of BU.

The SRM consists of a resolution authority (or Single Resolution Board – SRB) and a Single Resolution Fund (SRF). The SRB became operational in January 2015, but it started to work at full capacity one year later, on 1 January 2016, the date when the SRF was also on the schedule. The Finance Ministers of the member states have decided to keep some elements of the functioning of the future SRF in the form of an intergovernmental agreement, which complements the SRM regulation. According to the terms of reference of the agreement, the fund is to be financed by bank levies raised at the national level. As a general rule, banks taking higher risks will pay higher contributions. Contributions, initially consisting of national compartments, which will be progressively mutualised and eventually merged into a single fund administrated by the Board, start with 40% of resources in the first year. National compartments would cease to exist when the fund reaches the target funding level of 1% of covered deposits in the participating member states or after an eight-year transitional period – i.e. by 2024.

Under the SRM Regulation, the SRB is required to calculate the contribution from each individual bank to the SRF each year. Contributions are determined by applying the method detailed in a Commission delegated act and the specifications provided for in a Council implementing act, adopted respectively on 21 October and 19 December 2014. The establishment of the SRF will thus entail a shift from national to European resolution, which has the implication that each member state’s banking sector will progressively contribute more to the European resolution fund with respect to what they will be contributing to the national fund under the BRRD. This is summarised in Fig. 2.

European Banking Union, Fig. 2
figure 566figure 566

Evolution of phasing-in (−out) of contributions to SRF (from national target levels in accordance to the BRRD) (Source: ECB (2015) data)

Full size image

The SRF has an overall target level of €55 billion. While this amount may seem small in principle – given the need to signal to the markets that a reliable backstop exists (see also Macchiarelli 2014; Gros and Schoenmaker 2014) – one should consider that the fund has been given the ability to borrow directly from the market, if decided by the Board (ECB 2015); the terms and conditions of this have not been disclosed yet. Secondly, explicit provisions for bailing-in exist, as detailed by the revised BRRD (SRF website, https://srb.europa.eu/en/content/bail (accessed August 2016)). Bailing-in would apply until at least 8% of banks’ total assets had been used. After this threshold, the resolution authority may grant the bank the right to use the resolution fund, up to a maximum of 5% of the bank’s total assets. Some have observed how the actual procedures for bailing-in may not only risk cutting credit in already fragile economies, but could also reduce the willingness of lenders to extend new credit, having overall a negative effect on the financial conditions of that country.

Bank contributions to the SRF began in January 2016. However, a plan on bridge financing was put in place in the context of the Five Presidents’ Report (Juncker et al. 2015) in order to avoid a situation in which the SRF would run out of monies while bank contributions were being consolidated. The agreement, which was reached by the Council of Ministers in December 2015, introduced public support through the establishing of national credit lines that would provide a loan to the SRF in the case of capital shortfalls before 2024. As well as providing support where needed, the establishment of credit lines is intended to enhance the standing of the SRF. Importantly, a common backstop to the SRF itself should follow before the end of the transitional period, as a last resort measure, in order to ensure the durability of the BU project as a whole, as the Five Presidents’ Report also recognises (Juncker et al. 2015). (See also Communication from the Commission to the European Parliament, the Council, the European Central Bank, the European Economic and Social Committee and the Committee of the Regions, ‘Towards the completion of the Banking Union’ COM (2015) 587. Available at http://eur-lex.europa.eu/legal-content/EN/TXT/?uri = CELEX:52015DC0587.) This will be difficult to achieve in the short term as it will require a more far-reaching fiscal agreement among the member states (for an extended discussion see the following section).

Banking Union and Fiscal Backstops

While it is widely recognised that the proposals and compromises reached to deal with deposit insurance and resolution represent an exceptional step forward, many member states underscore that a well-functioning BU will require an unlimited burden-sharing mechanism, where fiscal authorities have to be involved. As highlighted above, the current design of the BU still leaves a role for an intergovernmental agreement, particularly in deciding the role and functioning of the future SRF – as a complement to the SRM regulation – before its final consolidation by 2024. Furthermore, the Commission’s deposit insurance mechanism (EDIS) is still on the negotiating table. Hence the stage in the governance framework that is lacking is the fiscal backstop.

The existing national DGSs and resolution funds – before a common backstop is created – may quickly run out of money and need last-resort support from sovereigns. This, however, was the origin of the so-called doom loop, pushing even countries with a sound fiscal record into a wrecking spiral, as the cases of Ireland and Spain show. Should this be the case, the sovereign will then need a backstop itself. In Fig. 3a, b we have used the ‘doom loop’ representation of Fig. 1 to summarise this discussion. The figure particularly compares (a) the current state of BU with a representation of (b) fully fledged BU in the context of the GEMU.

European Banking Union, Fig. 3
figure 567figure 567

European banking union. (a) Current state of European banking union; (b) European banking union in the context of the GEMU – in theory (Note: The figures include the main reforms of the European economic governance framework already in place (green); measures to be adopted during the transition to a BU (orange), and measures not yet in place (red). They do not consider measures which are temporary in nature such as unconventional monetary policy)

Full size image

The current state of BU is an incomplete banking union which could create coordination failures and be costly overall (Posen and Véron 2014). As mentioned above, leaving resolution funding and safety nets predominantly at the national level or, equally, limiting the BU’s ‘federal’ reach and burden-sharing capacity, would mean perpetuating the bank–sovereign doom loop, which is what the BU is intended to break. Alternative arrangements exist, but it remains to be seen whether these are convincing.

Figure 1(a) highlights a role for the European Stability Mechanism (ESM), as a last resort support. The ESM was primarily created with the purpose of providing a fiscal backstop for member countries and (more recently) their banking systems. However, the stability of banking systems can be assured only if investors know that such a backstop is not limited ex ante. This is why many commentators have defined the ESM as a ‘poor surrogate’ for a last-resort lender (De Grauwe 2011). The main reason why the European banking sector needs a common backstop is fundamentally macroeconomic and has to do with the very nature of systemic risk (Gros and Schoenmaker 2014; see also Allen et al. 2011). Once all of the above is in place (SRF plus EDIS), in the great majority of cases no public support will be needed. But in exceptional circumstances, for relatively large shocks, additional resources might be necessary, and clear arrangements on backstops should be made. Thus, the stability of banking systems can be assured only if investors know that such a European backstop exists, and the current ESM capacity can hardly be credible (see also Gros and Schoenmaker 2014).

Secondly, there are agency costs to consider, as the ECB/SSM may itself be trapped in a fiscal dominance game (see also Goodhart and Schoenmaker 1995; European Parliament 2012). The existence of a transition period before the SRM and the EDIS (whose deadlines for full functioning are aligned) makes it possible that, until resources are fully mutualised, the SSM will have an incentive to offload the fiscal cost of any problem to national authorities if it thinks that any given bank is insolvent and needs to be restructured or closed down. The SSM would do this on the basis of its comprehensive assessment of the viability of the bank and any danger it might constitute for financial stability. By contrast, national authorities in charge of bank restructuring would have a tendency to minimise their own costs by keeping the bank (even if illiquid) solvent through ECB support. This leaves some grey areas in the crisis management capacity of the BU (ECB 2015), with this type of conflict being prevalent between now and the start of the new system, when mutualisation is low. The endgame would be accelerating the process of consolidation of European deposit insurance and resolution schemes, thus minimising potential costs and avoiding providing the SSM and national authorities with the wrong incentives. (Other inter-agency conflicts and fiscal dominance may arise in the context of keeping two different coffers for European deposit insurance and resolution, respectively (for an extended discussion see Gros and Schoenmaker 2014).)

The nature of fiscal backstops beyond resolution and safety nets will be a crucial issue to define in the coming years.

Safety Nets

Authorities are now equipped with a broad set of tools to ensure that the costs of bank failure will, in the first instance, be allocated to bank shareholders and creditors following a clearly defined hierarchy (bailing-in), and only later involve dedicated resolution funds held at the national level (bailing-out). (Higher coverage will be granted for deposits related to certain transactions (e.g. real estate transactions and payment of insurance benefits). See ECB (2015).) In particular, as far as deposit protection goes:

  • Citizens’ covered deposits up to €100.000, representing about 48.6% (47.3%) of total euro area (EU) deposits, will be exempt from any loss. This number goes up to 70.9% (66%) for the euro area (EU) if the eligible over total deposits ratio is considered (author’s computation from Cannas et al. (2014) data).

  • Deposits of natural persons and SMEs above €100.000 will benefit from preferential treatment (they will not suffer any losses before other unsecured creditors do).

The Deposit Guarantee Schemes Directive (DGSD), which was transposed by the member states into national law in July 2015, concentrated on harmonising existing national deposit guarantee schemes without any common funding element. While regulators agreed to an increase in the minimum coverage of insured deposits from €20,000 to €100,000 and an increase in the speed of repayment for insured depositors, the most worrying gap is that of the unification of deposit insurance within the banking union.

In November 2015 the Commission put forward a legislative proposal to fill this gap, i.e. a European deposit insurance scheme (EDIS), taking a concrete step towards completing the third leg of BU. This is a very significant proposal, as the absence of a union’s deposit guarantee that could credibly back it underscores the dangers of incompleteness. A DGS funded at the European level would, in this case, make a material difference because it would provide an external loss absorption device that would be independent of the fiscal position of that sovereign (Posen and Véron 2014). Yet the EDIS has still to be approved, and it is currently stalling owing to political opposition.

The DGSD stipulates new thresholds for the financing of the national Deposit Guarantee Scheme (DGS), notably by requiring a significant level of ex ante funding (0.8% of covered deposits – or, where viable, a target level of 0.5% of covered deposits for highly concentrated banking systems) to be built up by 2024 by each member state. By that date, the Commission’s proposal envisages that resources will be mutualised in the EDIS. With the EDIS, the protection of deposits would be fully guaranteed at the European level, supported by close cooperation with national DGSs (Fig. 4). Given that national DGSs may remain vulnerable to idiosyncratic shocks, the purpose of EDIS would be to ensure equal protection of deposits in a centralised manner.

European Banking Union, Fig. 4
figure 568figure 568

Evolution of EDIS and participating DGSs funds in the Commission’s proposal (Source: European Commission website – European deposit insurance scheme)

Full size image

In the period elapsing between now and the EDIS, harmonised national deposit guarantee schemes will be necessary, meaning that member states concerned by a particular resolution plan will have to provide bridge financing from national sources (ECB 2015). In particular, if capital shortfalls are identified, the Council clarified on 15 November 2013 the order of the backstops. In the case of insufficient ex ante funds, the DGS must collect ex post contributions from the banking sector. Exceptional contributions should not exceed 0.5% of covered deposits per year. In the first instance, banks will thus have to raise capital in the market or raise capital from another private source. Should this not be sufficient, public money could be engaged at the national level in line with state aid rules and, if needed, through the provision of public backstops. Here, the DGS may have access to alternative funding arrangements, such as loans from public or private third parties. The DGSD also establishes a voluntary mechanism of mutual borrowing between DGSs from different EU countries (ECB 2015), the viability of which has still to be tested, particularly given the possibly competing interests of debtor and creditor countries.

Should national backstops not be sufficient, instruments at the European level may finally be enabled, including the ESM, consistent with the ESM’s agreed procedures. On the latter point, the Eurogroup agreed that the ESM would have the possibility to recapitalise ‘systemic and viable’ banks directly, with maximum exposure for direct bank recapitalisations capped at €60 billion (equal to 12% of the ESM’s maximum lending capacity).

Given the uncertainty about the full viability of the project in the medium to long term, information to markets and depositors should be prepared and coordinated.

Managing ‘The Outs’

One issue with the current approach to the European BU is that it minimises the importance of cross-border externalities of bank failures across the EU. Given the skewed design of the BU towards the euro area member states, the problem of funding is likely to be more severe when it involves guarantees to or resolution of banks which are systemic in both euro area and EU-non-euro area countries. For that reason, some of the ‘outs’ may make use of their option to opt-in to BU going forward (Gros and Schoenmaker 2014), provided that European resolution and deposit insurance schemes will be available. In this respect, the UK’s vote to leave the EU will place both the EU and the UK in uncharted waters, given the large presence of important European banks in London, and in the absence of clear rules on cross-border banking supervision and resolution under BU across EU and non-EU member states.

Final Remarks

A European banking union centred on the idea of single supervision, single resolution and a common deposit insurance system may be the most far-reaching reform to date since the inception of the euro (Constâncio 2013), if successful. Overall, however, the political resistance of creditor countries may restrain the effectiveness of crucial elements of BU, such as resolution and safety nets. A credible banking union would entail moving responsibility for potential financial support from the national to the supranational level, implying transfer of resources and risk, and, henceforth, requiring an explicit agreement on fiscal European support in the longer term. The latter agreement is a necessary step in the broader context of the European governance framework (see Genuine Economic and Monetary Union), in particular in achieving long-term ‘sustainability’ (see also Pisani-Ferry 2012; Gros and Schoenmaker 2014; Posen and Véron 2014; Schoenmaker and Véron 2016). For the time being, political resistance mainly focuses on the issue of permanent and unlimited vs. temporary and limited burden-sharing, leading to a ‘small steps’ approach.

An incomplete banking union can create coordination failures and could be costly overall (Posen and Véron 2014). An incomplete union can be interpreted in two ways. One is that it is a sequence in which much remains to be settled, but with reasonable clarity about the eventual destination. In this interpretation, the principal policy challenges will be how to manage the transition until 2024. The alternative explanation is that political resistance to burden-sharing will mean that only an incomplete banking union can be attained in fact. As mentioned above, leaving resolution funding and safety nets predominantly at the national level – i.e. the current state of BU – or, equally, limiting the BU’s ‘federal’ reach and burden-sharing capacity, would mean perpetuating the bank–sovereign doom loop; which is what the BU is intended to break. The nature of fiscal backstops beyond resolution and safety nets will be a crucial issue to define in the coming years.

See Also