1 Introduction

Ladies and gentlemen

I hope you have returned from the lunch-break suitably relaxed and refreshed, as I would now like to take you on a tour of a major construction site. But to put your mind at ease straight away, we won't be going to the Berlin airport or to the "Stuttgart 21" project – we don't have time for that in any case.

What I have to say concerns the construction work on a large-scale project of pan-European dimensions – the banking union.

2 Two pillars of banking union already erected ...

The first construction phase of the banking union was completed in November 2014, when the European Single Supervisory Mechanism, or SSM for short, was launched. Since then, the largest banks in the euro area have been supervised according to uniform standards. Certainly, other large-scale projects could learn a thing or two from the speed at which the SSM was put into operation. Of course, there are still some construction defects and design faults that need to be rectified. My colleague Sabine Lautenschläger highlighted some of them in her speech this morning.

And the second part of the construction, too – the European Single Resolution Mechanism (or SRM) – has been in place since the beginning of this year. Some of you will recall that Joanne Kellermann, a member of the Single Resolution Board (SRB), spoke here a year ago and gave you an early introduction to the SRB's responsibilities.

The SRM is an important complement to common European supervision. Its purpose is to ensure in a worst-case scenario that if an institution stands no chance of survival, its shareholders and creditors will be first in line to bear the ensuing losses. Then, a single resolution fund that is well stocked with bank levies can play its part, meaning that taxpayers would be at the very end of this liability cascade. In this way, the European Single Resolution Mechanism also performs a role in severing the close links between banks and public finances which, as you all know, were a key problem during the crisis.

These two pillars of banking union have given a tremendous boost to European financial integration. But there is still a considerable need for debate on the matter of a third pillar – the European deposit insurance scheme, or EDIS for short.

3 ... and the third is yet to come

This topic has been the subject of heated debate, especially in Germany, since November of last year, when the European Commission drew up bold plans for the design and timeline of its EDIS proposal. And it's no surprise that emotions are running high.

The objective of the European deposit insurance scheme is as old as deposit insurance itself. It is to prevent a run on banks by protecting deposits. To achieve this requires two things: a functioning protection fund and a binding legal framework.

At first sight, the European Commission's proposal appears to be very logical. In a European integrated financial market, the risk of a run on banks needs to be minimised at the European level, too, by safeguarding deposits. If a purely national deposit protection system got out of its depth, a European scheme would come to the rescue.

What does the Commission's proposal envisage in detail? The first phase, due to run from 2017 to 2019, will take the shape of a reinsurance system consisting of national compartments, so to speak, with a European reinsurance fund as a backstop. This would be followed, from 2020 to 2023, by a co-insurance phase, where the national scheme and the European fund pay from the very first euro on in the event of an insolvency. The European fund's share would increase steadily until the third phase kicks in. By 2024 at the latest, a system of comprehensive insurance would be put in place at the European level. From then on, all compensation events occurring in the participating countries would be covered by the European fund.

To be frank: it is not unlikely that we will one day see a Europeanisation of deposit insurance as the third pillar of banking union. We have to fulfil the necessary conditions for that now. And then we need to think about how a third pillar can best achieve the objective in mind – namely, to protect deposits, thereby preventing a run on banks.

Those of you who have built a house will know that there's more to it than drawing up bold plans: the fundamental conditions also have to be taken into account, such as the nature of the ground and the building area or structural issues. And as far as the fundamental conditions are concerned, a great deal of work still has to be done before we can embark on the third construction phase.

4 No foundation, no pillar

You may have heard about the time Sherlock Holmes and Dr Watson went camping. After a delicious repast and a bottle of wine they retire to their tent to sleep. Holmes wakes up a few hours later and promptly rouses his friend. "My good Watson," says Holmes, "kindly direct your gaze upwards and tell me what you see." Watson looks up and replies: "I see millions of stars, Holmes." "And what, pray, do you infer from that?" Holmes asks. After a moment's reflection, Watson answers: "In terms of astronomy, that there are millions of galaxies out there, and possibly trillions of planets." […] In terms of the time, I would say it is roughly a quarter past three. […] Meteorologically speaking, I would conclude that we are in for clement weather tomorrow. And what, pray, do you infer?" Holmes' eyes flash momentarily as he says: "You fool, Watson, someone has stolen our tent."

Ladies and gentlemen, my reason for telling you this story is this. Like our expert sleuth and his trusty sidekick, the architects of the common deposit insurance scheme are well-advised to separate the crucial from the merely incidental. And our prime objective is to protect the bank customers' deposits. Whether or not this necessarily has to happen at the European level at this point in time is, in my view, of secondary importance.

Instead, I would argue that making the third pillar of banking union as stable as possible hinges on three factors. First, the necessary preconditions have to be fulfilled; second, we have to think very carefully about whether the Commission's proposal is the best way to achieve our objective; and third, tried-and-tested national systems cannot simply be discarded. Allow me now to address these three factors in detail.

Let's begin with the necessary preconditions. One major element when we speak about safeguarding deposits is stable banks, for they offer the best possible protection for deposits. And Europe still has much to do in this respect. In my opinion, this includes, first and foremost, the following.

First, risks at the European level can be communitised only if all the members of the common deposit insurance scheme make the same efforts to limit the risks. To do this, it is essential that all parties implement and comply with the existing rules. Above all, this means implementing the agreed measures for the restructuring and resolution of credit institutions as well as those for harmonising the existing deposit protection schemes. Unfortunately, not all the member states are meeting this objective at present. And these rules do not just have to be implemented, they also have to applied effectively. This requires an adequate buffer of bail-inable liabilities to be in place, for example, as well as legal certainty about the possibility of a bail-in. This will necessitate follow-up adjustments in terms of treatment under insolvency law.

Second, my concern is with dissolving the bank-sovereign nexus: many banks in the EU have considerable claims on sovereigns on their balance sheets, especially on the government of their own country. Naturally enough, this involves matching credit and concentration risks. But as long as the holding of sovereign bonds enjoys preferential regulatory treatment, banks will have no incentive to reduce such risks on their balance sheets and will thus remain dependent on the economic situation in their home country. If the status quo were to remain in place, a single deposit insurance scheme could mean the communitisation of sovereign debt through the back door.

It is my firm belief that such a political temptation to hijack the single deposit insurance scheme for another purpose has to be avoided at all costs. It is therefore gratifying that the European Commission, in its proposal on a European deposit insurance scheme, explicitly goes into the problem posed by the preferential regulatory treatment of sovereign bonds. But now the Commission has to back that up with a concrete proposal which addresses both credit and concentration risk. I believe that a timely solution to this problem will be crucial for the ongoing negotiation on a deposit insurance scheme.

And I would like to state yet another prerequisite for a functioning deposit insurance scheme: striking a balance between action and liability. European liability should exist only if the European level also possesses sufficient rights of control. And we are still a long way from that.

Despite European banking supervision and resolution, national economic policy still has a very big influence on domestic banks’ economic situation. The same goes for the underlying legal conditions. At the national level, there still exist very major differences in the law pertaining to insolvency, for example. Naturally enough, the fact that such rules differ nationally has direct implications in terms of the banks’ risk position and the burdens that lie in wait for them in the event that their borrowers become insolvent. This means that they also have an impact on how likely the banks are to draw on the deposit insurance scheme ‒ and how heavily.  So, the possibility of this creating unfair conditions makes stronger European action in national economic policy as well as harmonised legal principles a fundamental requirement for a European deposit insurance scheme.

Constructing a European deposit insurance scheme can make sense. But for it to really fulfil the principal task intended for it by the architects of banking union, we first have to repair the irregularities and gaps in Europe’s financial system.

What should be crucial in all deliberations is ensuring the best possible protection for bank customers’ deposits. And, looking at the conditions that are in place, it isn’t immediately obvious at the present moment why that should function better at the European level than nationally.

I say that because the picture at the national level is not at all as dramatic as it is painted by the proponents of the European deposit insurance scheme. You see, as a result of the revised European Directive on Deposit Guarantee Schemes, which has been in force since mid-2014, we already have a harmonised system in Europe. Admittedly, it is a system where the funding schemes are national, but which has to adhere to minimum European standards. These standards provide for all national funding schemes having to save a target volume of 0.8% of covered deposits. This funding makes it possible to meet the statutory obligation to protect deposits per customer and per credit institution up to the coverage level of €100,000. This was a major step forward, and our first priority should be to get this system up and running.

5 Using the scope for shaping the third pillar

Ladies and gentlemen, in the long term we shall not be able to shut ourselves off from a debate on a European, Community solution ‒ of that I am absolutely certain.  But before we can celebrate the completion of the banking union structure, there is still a great deal to be done. And that raises the question of whether the European Commission’s current architectural blueprint is also the best design possible.

Let us return to Watson’s problem in terms of not losing sight of what is crucial.  The objective of a deposit guarantee scheme is always the protection of deposits and thus preventing a run on banks. In the vast majority of cases, national deposit guarantee schemes are adequate for that purpose. Only when a systemic crisis arises, putting the entire financial sector at risk, is a national scheme unable to cope.

How well is the Commission’s proposal suited to achieving that objective? Do better alternatives exist? Is the EDIS proposal really the most efficient way to construct a third pillar to bear the European burden?

Not everyone seems to think so. Indeed, some distinguished economists believe that a European reinsurance system would be just as effective – and more efficient to boot.[1] A reinsurance scheme would install the national funds as the first lines of defence, with a European reinsurance fund as a backstop. This approach could be a way of leaving sole responsibility for tackling purely national cases – and the vast majority of them will probably be national cases – with the national funds, once and for all. Under this set-up, the European component would be an insurance backstop for the national deposit protection schemes, should the worst come to the worst. It would be equally conceivable to do away with a European fund altogether and permit mutual lending among the national systems. That's pretty much what's already sketched out, albeit in non-binding terms, in the existing Deposit Guarantee Directive.

Of course, that kind of system shouldn't come into effect before the preconditions I mentioned have been met, especially the ones concerning de-risking. I'm not sure whether this is the right path to take, so I will reserve my judgement, but it is at least worth mentioning as an alternative.

Both approaches – the reinsurance fund and the intergovernmental agreement – would ensure that problems that can be fixed nationally remain in national ownership; systemic crises that overwhelm countries would be a matter to be addressed by the European reinsurance backstop. I firmly believe that these two approaches – and other possible solutions besides – should be scrutinised very closely indeed before construction work kicks off. We should let our imagination run wild here, as long as we don't hastily mutualise risk.

6 Tried-and-tested protection schemes worth preserving

All these deliberations make it quite plain that the Commission's proposal isn't necessarily the be-all and end-all of the matter and that if we want to bolster the European deposit insurance component, we should devote some thought to other possible solutions as well. All the more so since I see no sign that the Commission has done its duty of analysing the impacts and alternatives upfront.

One factor, I feel, is particularly crucial here. It would be wrong to simply discard tried-and-tested national systems. A workable safety net for the financial sector doesn't necessarily need to be a purely European brand. A safety net is workable if it makes full use of every single robust instrument to prevent customers from being locked out of their bank.

In some respects, the existing deposit insurance schemes in Germany and Europe go far beyond the mere payout function which the Commission is lining up for the EDIS. Indeed, the institutional protection schemes in the savings bank and credit cooperative sectors reinforce the safety net if one of their members is on the brink of insolvency – I'm talking about institutional protection here. Similarly, the voluntary deposit protection offered by the private commercial banks makes a payout under the deposit protection scheme a less likely scenario.

Before I continue, let me make it clear that these agreements are not about, and cannot be about, circumventing the requirements of statutory deposit protection. Of course, the statutory deposit protection schemes run by German institutions, just like all the other European deposit protection schemes, are obliged to guarantee their customers' covered deposits up to an amount of €100,000 and to actually build up a volume of funds earmarked for this very purpose. This requirement is designed to ensure that voluntary preventive measures provide added value, and we shouldn't risk losing this added value.

There is good reason to be sceptical about initiatives put forward by the financial sector. What's in it for institutions that voluntarily do more than they are legally required to? Unsurprisingly, they stand to gain, too. Anyone charged with administering an association's mutual compensation arrangement is keen to find low-cost solutions if its members are teetering on the brink. Thus, an institutional liability arrangement encourages members to look a little more closely at how counterparties are running their businesses early on. The Bundesbank supports these voluntary protection schemes provided and as long as they work effectively. Whether that is assured very much depends on the extent to which cooperation is actively fostered and promoted in the associations.

How will the plans for an EDIS affect institutional protection arrangements? Legally, not at all, but economically, they will feel the heat. That's because the draft EDIS requires all institutions, every single one of them, to pay into the European insurance fund. In other words, if institutions are members of a protection facility that makes a payout event less likely, they will effectively be paying into a fund they might never need to tap in the future. This has raised concerns, especially in the savings bank and credit cooperative sectors, that the European insurance scheme will undermine the balance between what institutions contribute and what they stand to get in return. Those are concerns I quite understand.

We need to make sure we don't lose any ground. We need to preserve the added value which the institutional protection schemes and voluntary deposit insurance arrangements deliver. If it is our intention to create the same self-supervision incentives under a European deposit insurance scheme as well, we now need to put well-considered economic ideas on the table. One possible option could be to suitably incorporate the actual risk of participants in an additional protection facility defaulting into the calculation of their contributions to the EDIS. Questions like this continue to have a major bearing on how well the planned third pillar of banking union will slot into the existing mesh of institutional safety nets.

7 Conclusion

Ladies and gentlemen, as you can see, there are different ways to protect European deposits. For all the points that speak in favour of a European deposit insurance scheme, there are also good reasons not to implement the Commission's proposal right now in its current form.

Before we can arrive at a European deposit protection arrangement, I see three problems that we urgently need to resolve once and for all.

• First, the agreed recovery and resolution measures for credit institutions will have to be implemented throughout the EU before we start discussing a possible third pillar of banking union. And the bail-in tool, being a core component of the new resolution regime, will need to be both legally and effectively feasible.
• Second, sovereign risk in bank balance sheets will need to have been reduced once and for all, and the preferential regulatory treatment of government bonds done away with.
• And third, Europe needs to make genuine progress in economic policy integration, including a general insolvency regime.

What is more, we shouldn't be making any momentous design decisions without first having looked in very great detail at realistic alternatives that might even offer a superior outcome. Lastly, we need to select the most durable structural design to make sure that the deposit insurance scheme stands firm in the event of systemic crises.

Tried-and-tested national systems ought to feature in a solid structural design of this kind – that's something we at the Bundesbank advocate.

Thank you very much for your attention.

Footnote

1. D Gros (2015) Completing the banking union: deposit insurance. CEPS Policy Brief No 335; D Schoenmaker, G Wolff (2015) Options for European deposit insurance. Comment on VoxEU, 30 October 2015.