The public and the private banking union
Speech by Jörg Asmussen, Member of the Executive Board of the ECB,
at the joint conference “The Single Resolution Mechanism and the Limits of Bank-Regulation”
Humboldt Universität/Financial Risk and Stability Net-work in Berlin
Berlin, 8 November 2013
Dear Mr Aehling [Veranstalter]
Ladies and gentlemen,
The timing of today’s conference on “The Single Resolution Mechanism and the Limits of Bank-Regulation” could not have been better chosen as the SRM is the dominating theme in all European fora at the moment.
Thank you for inviting me and I am happy to share a view on banking union from the European Central Bank with you.
Looking back with the experience of the crisis, I think that many here today share the assessment that the key problem in the past was that there was too little cooperation and too much diversity – we lacked a coherent European approach to financial sector policies. But there is clearly scope for more cooperation today as, particularly after the experience of the crisis, preferences are more aligned in Europe. We all want banks that are well supervised and that can be resolved at minimum cost to the taxpayer.
To achieve this, I think we need not reinvent the wheel. There are elements of the US approach that have worked and that we can usefully borrow in Europe. Saying this, one has to acknowledge, of course, that the US is a fully integrated federal state whereas the European Union will remain a construction sui generis for the foreseeable future. Therefore, the US approach cannot serve as a blueprint for Europe, but its direction may prove right for Europe as well.
The elements of the US approach fall into two basic categories: first, what I call the US’s ‘public banking union’ – its federal institutions for dealing with banking crises. Second, what I dub its ‘private banking union’– its integrated financial market that dissipates the costs of crises.
Public banking union
Clearly, the key difference between Europe and the US is that banking problems in the US have been mainly taken care of at the federal level, whereas in Europe responsibility for troubled banking has thus far remained national.
During the crisis in the US, it was federal institutions that stepped into the breach. The Federal Deposit Insurance Corporation (FDIC) resolved 490 unviable banks, using resources from its deposit insurance fund financed by all the banks. The US Treasury provided recapitalisation support and guarantees through its Troubled Asset Relief Program (TARP). The Federal Reserve undertook a thorough stress-test, determining banks’ additional capital needs.
To understand the situation we are in in Europe, it is useful to make a thought-experiment, and imagine that no such federal institutions exist in the US. Imagine that the state of California would have had to come up with the TARP money for Wells Fargo instead of the US Treasury. Imagine that the state of New York would have had to provide the guarantees to Citigroup. Think about how the markets, how would the rating agencies have treated these states? And how they would have treated these banks? The lesson for the euro area is clear: federal institutions can prevent local crises from becoming systemic.
In Europe, we do not yet have these federal institutions that can act as shock absorbers. So far, responsibility for dealing with bank problems lies exclusively with the individual EU countries. The result is that – to quote an FT editorial – “national taxpayers subsidise banks implicitly in good times, explicitly in bad times, and suicidally in a sovereign debt crisis”.
Indeed, the crisis in Ireland has shown that problems in domestic banks can overwhelm the fiscal capacity of national sovereigns. The causality can also be the other way around, as the case of Greece has shown, where fiscal problems have dragged down the banking system. These countries got caught in a vicious fiscal-financial feedback loop, which in the end also drags down the economy, breaks up the internal market, and impairs our monetary policy conduct. Removing this vicious feedback loop is what Banking Union is all about.
Creating a banking union will be good for the euro, good for the internal market, and good for the recovery. It will be good for the euro, because it will help to repair our monetary policy transmission mechanism.
In other words, it will ensure that the ECB’s low interest rates will be effectively passed on to those countries that probably need them most. It will be good for the internal market, because it will be of help to reverse the significant financial fragmentation that we have seen so far. And it will be good for the recovery, as banks will be put on a stronger footing, confidence will be improved, and ultimately credit will start to flow again to enterprises on affordable terms.
But what are the necessary elements of a banking union? To our mind, two elements are indispensable: a Single Supervisory Mechanism (SSM) and a Single Resolution Mechanism (SRM).
Furthermore, it is important to have a single rulebook and the implementation of the Capital Requirements Directive and Regulation IV is key in this respect. A single European deposit insurance is not top on the regulatory agenda today and it may be tackled in the long term only. In fact, the Bank Recovery and Resolution Directive (BRRD) and its liability cascade makes it clear that deposits will come in only at the very end in bank resolution and thus the need for a single deposit insurance scheme is less urgent and a better harmonisation could suffice at present.
Let me start with the Single Supervisory Mechanism. The ECB has been asked to assume responsibility for this mechanism, and to mould the existing national supervisory authorities into an effective and efficient federal supervisory team. We are making good progress on this front, and the SSM should become fully operational in the last quarter of 2014.
Before we take on this task, we want to get an adequate view of what we will be supervising. In the coming months, the ECB will therefore undertake a balance sheet assessment.
The importance of this exercise cannot be overestimated. This comprehensive exercise will encompass three steps:
First, a supervisory risk assessment to review, quantitatively and qualitatively, key risks, including liquidity, leverage and funding.
Second, an asset quality review (AQR) to enhance the transparency of bank exposures by reviewing the quality of banks’ assets, including the adequacy of asset and collateral valuation and related provisions.
Third, a stress test to examine the resilience of banks’ balance sheet to stress scenarios.
These three elements are closely interlinked.
The comprehensive assessment will conclude with an aggregate disclosure of the outcomes, at country and bank level, together with any recommendations for supervisory measures. This comprehensive outcome will be published prior to the ECB assuming its supervisory role in November 2014, and will include the findings of the three pillars of the comprehensive assessment.
We will not publish any preliminary or intermediate results and I am quite surprised about the noise you hear these days in all directions about the possible outcome of the exercise. If we knew that “banks in the periphery will not face severe problems” or if we knew that “the recapitalisation needs will be a double digit bn figure” we could spare the effort in conducting this exercise. All these statements are mere speculation.
The comprehensive assessment is an extraordinary opportunity to rebuild confidence in the banking sector and promote the long overdue repair of bank balance sheets, which in turn is a precondition for a sustainable economic recovery.
One particular point needs to be stressed: credible national backstops must be put in place. If not, the credibility of the whole exercise is put at risk as the outcome will then almost certainly be negatively perceived by market participants. Doing this balance sheet assessment without a backstop in place would be a bit like getting on a boat in rough weather conditions, and not taking a life jacket on board. Any recapitalisation needs uncovered by the exercise should of course first and foremost be covered by the market. National budgets and national resolution funds may intervene as a second line. Finally, a European backstop, the ESM with its existing instruments, meaning a banking sector programme like in the case of Spain, may be available.
The second key element of a banking union is the Single Resolution Mechanism, a European equivalent to the Federal Deposit Insurance Corporation. The crisis has shown that simple coordination among Member States has been inadequate to organise the resolution of cross-border banks in an efficient, quick and least-cost manner.
The Single Resolution Mechanism contains three essential elements:
First, a single regulatory framework. This will be established with the Bank Recovery and Resolution Directive (BRRD). The trilogue of the Council, the European Parliament and the Commission is currently ongoing and from an ECB perspective a desirable trilogue outcome would contain two elements.
First, bringing forward bail-in to 2015 as markets will anticipate bail-in anyhow. Having the rules in place earlier will give investors’ bigger security.
Second, the flexibility and scope for national solutions should be somewhat reduced so that market participants will have sufficient confidence about the rules of the game in Europe. The second element of the single resolution mechanism is the single resolution authority and the third, a single resolution fund, funded by contributions from the financial industry. The last two elements are currently in the legislative process and we strongly support the envisaged timeline for their entering into force (1 January 2015), as it ensures that the establishment of the Single Resolution Mechanism would follow shortly after the effective start of the Single Supervisory Mechanism.
Private banking union
The ‘private banking union’ in the US is created by its integrated financial market, which includes both the banking sector and capital markets.
In the banking sector, the US has genuinely nationwide banks, which allows them to better diversify risk. If such a bank makes losses on loans in one state it can offset them with gains in another – thus at least in part internalising the costs. In the euro area, however, banks tend to have a more domestic focus, meaning losses in one country are more likely to exhaust the capital of the domestic banks.
In the capital markets, the US has integrated equity and debt markets, allowing losses from bank failures and corporate insolvencies to be dispersed across the whole market. In the euro area, by contrast, cross-border capital markets are only partially developed, which means the effect of a negative economic shock is fully concentrated at home. In Germany and France, for example, around 85% of equity is held domestically.
One should not underestimate the importance of this: it has been estimated in a paper that two-thirds of economic shocks in the US are absorbed via the integrated financial market.  While the launch of the euro seems to have encouraged more private risk-sharing between EMU countries, it remains at much lower levels than between US states, and markets are vulnerable to fragmentation. 
This point underscores that the private and public banking unions are to a significant extent interdependent. Having a strong set of federal institutions can stop market failures such as a fragmentation during a crisis. Similarly, having a single financial market can make the task of federal institutions easier, in particular for resolution, as it allows mergers and takeovers to be used as a primary resolution strategy.
Indeed, 450 out of the Federal Deposit Insurance Corporation’s 490 bank resolutions since 2008 have involved what is called “Purchase and Assumption” – selling parts of banks to other banks, often located in different states.
The conclusion from this is simple: completing the banking union and completing the single financial market go hand in hand.
Ins and Outs
This brings me to the topic: how does the banking union relate to countries inside and outside the euro area?
I will make three points here.
First, the banking union is critical for the Euro Area countries.
Second, it is also desirable for Member States, which do not share the euro.
And third, the banking union should be about strengthening the internal market, rather than dividing it in two.
Let me start with the first remark: the banking union is critical for the euro area countries.
The reason why we are building it is to break the vicious circle between sovereigns and banks, the manifestations of which are much more acute and disruptive in a monetary union.
That is why we need the Banking Union in the single currency area. But one thing must be clear: putting the Eurozone on a stronger footing will also benefit the countries that are not in the eurozone. The crisis has clearly shown that in Europe a receding tide lowers all the boats, not just the ones which are in trouble.
My second remark is that the banking union – whilst critical for the ins – is also desirable for the outs. The so-called financial trilemma  – in other words the impossible trinity of financial integration, financial stability, and national responsibility for crisis prevention and management – holds for the entire region, and not just for the euro area. For example, the Swedish and Danish banking system are not only large in proportion to your GDP, but they are also very closely integrated with the rest of Europe.
Let me focus on the case of the Danish banking sector. Danish banks’ consolidated assets are almost four times Denmark's GDP.
The biggest banks have considerable cross-border activities. For these reasons, it would make perfect sense for Denmark to join a Banking Union that is composed not only of a single supervision mechanism, but also a single resolution mechanism with a resolution fund that can serve as an insurance scheme in relation to systemic banks.
Whilst participating in the banking union therefore appears desirable, the terms of that participation should also be right.
Indeed, we at the ECB consider it of the utmost importance to ensure that the participation of the outs occurs on equal terms as the Euro Area countries.
On resolution, we need to find creative solutions to ensure a level-playing field between ins and outs.
This holds also true for the backstop for the single resolution fund which is currently the subject of much discussion.
Taking part in the Banking Union from the beginning would bring many benefits. Having a chair at the table now allows participants to steer the practical implementation of the SSM – for example the development of the supervisory manual, guidelines and general instructions, as well as benefit from the ‘healing effect’ that the comprehensive assessment should bring.
Let me make a third remark: The banking union should be about strengthening the internal market, rather than fragmenting it.
Both the single currency and the single market are key pillars of growth and prosperity in Europe. Both should be maintained and enhanced. Hence, the banking union should not create new barriers within the internal market.
If countries choose not to join, they should not be discriminated against. Two safeguards will be crucial in this regard: having a single rulebook that sets the same standards for all, and having a strong and effective European Banking Authority with a fair representation of the outs.
The relationship between banking and fiscal union
Let me now turn to the relation between banking union and fiscal union.
For the fiscal union, finding the balance between cooperation and diversity is difficult. National preferences for taxing and spending are stronger, and the case for elevating certain functions to the European level is more complex.
And there can be no taxation without representation, meaning any shift to a more European fiscal policy must be accompanied by the same shift towards European democratic control.
So to frame the discussion, I think it is helpful to recall the taxonomy of economic policy functions described by Richard Musgrave. He distinguished between stabilisation, allocation and distribution. Broadly speaking, stabilisation helps smooth the business cycle, allocation ensures that resources are used efficiently and distribution ensures an equitable sharing of income.
Concentrating on stabilisation, there is a long-standing debate in Europe about whether monetary union needs a federal budget to support counter-cyclical stabilisation. As you may know, the report by the four Presidents on a genuine EMU suggested that a small fiscal capacity could be used for this purpose.
I do not want to comment on the specifics of this proposal, but there are questions about whether national preferences are sufficiently aligned to permit it. For that reason, we have to think carefully about whether it is essential for the euro area. In particular, we have to consider how having a genuine banking union might change the debate.
As you know, stabilisation in Europe takes place through national budgets and, in normal circumstances, they are sufficient to smooth the cycle. A recent study demonstrates that 47% of an unemployment shock is absorbed by the automatic stabilisers in the EU, compared with only 34% in the US.  They were insufficient in some countries during the crisis only because they faced an exceptional circumstance: a very large shock from the financial sector.
With a genuine banking union, however, the risk of this happening again should be lower. First, we expect the SSM to increase the quality of supervision and the assessment of systemic risk. Second, if a financial shock were still to appear, national budgets should be better insulated. The costs of banking sector repair would be shared with the private sector through the Single Resolution Mechanism and a more integrated European financial market.
All this suggests that stabilisation at the national level could be more effective in future. Or put differently, the more successful we are in building a genuine banking union, the less we would need a central budget for stabilisation purposes.
However, this does not mean I see no role for fiscal union in the stabilisation area.
First, we would need the public backstop, namely for the future resolution fund, which would be fiscally neutral over the medium-term. In my view the ESM could play this role like the US Treasury does this for the Federal Deposit Insurance Corporation by providing a credit line. Since there has been some confusion in the past, I would like to reiterate that this would indeed require a change of the ESM Treaty but not of the EU Treaty.
Second, we would need an insurance mechanism against tail events that may still arise and endanger euro area financial stability. Here we already have the ESM to act as lender of last resort to sovereigns.
To sum up: a functioning banking union reduces the need for the stabilisation function of a fiscal union, but not to zero.
I have tried to give you the ECB’s view on the various aspects of the banking union. Our preparations to take over the task of the SSM are well under way. We are adamantly convinced that banking union will only function if the SSM is complemented by a single resolution mechanism and that it is put in place without undue delay. Given the parliamentary recess as of May next year in view of the upcoming elections to the European Parliament, it is of utmost importance to have a political agreement on the Single Resolution Mechanism in the ECOFIN by the end of this year.
To make this happen, all need to move. The following are the main open issues that need to be solved:
First, the legal base: Article 114 of the EU Treaty on the establishment and functioning of the internal market is a possible legal base from our point of view, but it could also be based on the flexibility clause of Article 352.
Second, the scope: The Commission proposal foresees that all banks in countries participating in the SSM should be covered by the SRM as well. My personal view is that a compromise should be possible here. Why not start with the systemic SSM banks, ie the 130 for which the ECB will be the direct supervisor, and progress further over time?
Third, the resolution authority: It must fulfil three criteria. It has to be in a position to take decisions quickly, if needed during a weekend. That said, it is obvious that the Council cannot be the resolution authority. The resolution authority must be established quickly. That means there is no time to wait for a treaty change to establish a new institution. The new authority must take a European perspective on financial stability.
Let me be clear that I do not see the task of the resolution authority with the ECB. There should be a clear division of labour between supervision and resolution.
The fourth and last big open issue is how to finance the backstop for the single resolution fund as long as it is not sufficiently funded via levies from the banking sector. As mentioned before, I think the ESM including any non euro-area member states could provide a credit line to the resolution fund. That would guarantee a clear control over the resolution fund by the participating member states and their parliaments.
A functioning banking union is in the best interest of Europe and it is in the best interest of Germany.
B. Sørensen and O. Yosha (1996), “Channels of interstate risk sharing: United States (1963-1990)”, The Quarterly Journal of Economics, November, pp. 1081-1111
Y. Demyanyk, C. Ostergaard and B. Sørensen (2008), “Risk sharing and portfolio allocation in EMU”, DG Ecfin Economic Papers 334
Schoenmaker, D., (2011). "The financial trilemma ," Economics Letters, vol. 111, pages 57-59, April
A recent study demonstrates that automatic stabilisers absorb 38% of a proportional income shock in the EU, compared with 32% in the US. In the case of an unemployment shock 47% of the shock is absorbed in the EU, compared with 34% in the US. This cushioning of disposable income leads to a demand stabilisation of up to 30% in the EU and up to 20% in the US. See: M. Dolls, C. Fuest and A. Peichl (2012), “Automatic stabilizers and economic crisis: US vs. Europe; Journal of Public Economics”, Elsevier, vol. 96(3), pp 279-294.