1 Introduction

Professor Keller
Professors
Students at the Bundesbank’s University of Applied Sciences here in Hachenburg
Ladies and gentlemen

I would like you to join me today as I take a look at the state of the European banking union. The students among you can breathe a sigh of relief: I’m not here to discuss committee structures or test your knowledge of EU institutions, and there will be no quiz after I’ve finished talking. The question that I would like to explore is one that is of interest to us and the general public alike: does the European banking union – that “big new thing” in the history of banking supervision – actually work?

Truth be told, such a fundamental question shouldn’t really need any prompting. Nevertheless, I will address the question by examining four recent examples of institutions running into difficulties in the European banking sector – and, to my mind, what better place to illustrate such a topic than at the Bundesbank’s University of Applied Sciences?

To start, which cases am I referring to? First, Banco Popular Español, which was wound up in June 2017 under the new European resolution rules. Second, Banca Monte dei Paschi di Siena. It received EU approval for precautionary recapitalisation, a procedure under exceptional circumstances in which an injection of state capital does not result in an institution being deemed “failing or likely to fail” or in it bearing the repercussions of such an assessment. My third and fourth examples are very recent – just two months ago, two Venetian banks were liquidated under Italy’s national insolvency regime. So, within a short space of time, we have seen three different approaches to dealing with struggling banks in the euro area.

In exploring these cases today, I would like to take the opportunity to delve deeper into the functioning of the banking union and to draw a handful of initial conclusions. I want to focus on three topics in this regard:

• first, the work of the Single Supervisory Mechanism (SSM);

• second, the role of the Single Resolution Mechanism (SRM) and the national insolvency regimes;

• third, the current state of affairs in the European banking sector, which is not to be overlooked in the context of the banking union.

2 Crisis situations – uncharted territory for the SSM

It isn’t for nothing that ailing banks attract so much public attention. This is because, by their nature, a great deal of money is at stake. That’s precisely why declaring that an institution’s failure is imminent or inevitable – known in technical terms as “failing or likely to fail” – is the most powerful weapon in the banking supervisors’ arsenal. It equates to forcefully bringing an end to an institution as we know it. The all but inescapable consequence is either resolution under the new European resolution rules or insolvency proceedings under national law. In economic terms, the decision draws a line – at least temporarily – under the institution’s accounts. Losses arising from business activities now need to be realised. And the value of assets in a gone concern – that is to say, an institution that needs to be wound up – is almost always lower than that of those in a going concern. It’s not least for this reason that banking supervisors should not be overly hasty in classifying an institution as “failing or likely to fail”.

But there’s also an old saying that applies here: it may be far better for all parties concerned to make a painful break rather than prolong the agony. Needlessly perpetuating the situation could very well push up costs and losses. What’s more, “gambling for resurrection” – by which I mean focusing on a failing institution’s unlikely prospects of recovery, which could leave creditors and taxpayers footing an enormous bill if things come to a head – is not a viable option. There is no room for wishful thinking here – and that applies to both financial institutions and their supervisors.

All in all, one thing is clear: when it comes to the timing of “failing or likely to fail” verdicts, banking supervisors cannot simply arrange these around the premier league schedule. A lack of confidence and liquidity can force the supervisors’ hand at extremely short notice. In each individual case, every single detail needs to be scrutinised so that a well-thought-out decision can be made. In this regard, cooperation within the SSM and with the SRM has worked well in the cases I have mentioned.

The use of precautionary recapitalisation, which I touched upon at the start of my speech, also marks a step into uncharted territory for European supervisors. This is enshrined in the new European Bank Recovery and Resolution Directive (BRRD) as an exceptional measure that enables an institution to receive an injection of public funds without - as is generally envisaged - being deemed “failing or likely to fail”.

Precautionary recapitalisation is only approved under certain conditions. In particular, the institution must be deemed solvent, and the public funds may not be used to offset existing or anticipated losses. Supervisors need to carefully examine whether these requirements are met. Additionally, the European Commission has to decide whether the granting of public funds is compatible with EU state aid rules.

It’s a particularly sensitive issue because one of the fundamental objectives of the new resolution rules is to shield taxpayers: it should be a bank’s shareholders and creditors who are on the hook for unsatisfactory bank management and any losses that arise as a result, not taxpayers. Then again, European legislators have accepted a serious disturbance in the economy of a member state and risks to financial stability as grounds for government intervention to prop up solvent institutions. The stringent criteria outlined in the legislation do actually need to be fulfilled, however.

In other words, these conflicting aims make it clear that precautionary recapitalisation cannot be a blueprint for averting bank crises. Consequently, precautionary recapitalisation is defined in the BRRD as a purely exceptional measure. This is also my clear understanding of the term.

What is also clear is that, in each and every instance, shareholders and subordinated creditors need to be bailed in before the government coffers are opened. With regard to Italy’s Monte dei Paschi di Siena, what this means in concrete terms is that the institution’s capital base was first bolstered by €4.5 billion by converting subordinated debt instruments into equity, with the Italian government then injecting an additional €3.9 billion. However, the government will offer compensation to retail investors, thereby increasing its stake in the bank. The ownership structure will change as a result of the aid provided: legacy shareholders will lose out as their stakes are diluted. By contrast, the Italian government will own up to 70 % of the institution. To sum up, the state is providing assistance to an ailing bank and, as an investor, will be exposed to any potential future losses.

Looking beyond this particular case, one thing is clear: while the use of public funds will be made more difficult under the new resolution rules and will only be permitted subject to compliance with certain criteria, it has not been taken off the table completely. Yet the fact that government aid was provided to as many as three of the first four banks to run into difficulties cannot be in the interest of European legislators. Our resolution regime should be measured against how effectively it shields national taxpayers – and I firmly believe that this should also be the goal of national insolvency law as well.

Before I go on to talk about the actual resolution cases in greater depth, let me take this opportunity to draw three conclusions from the crises as viewed from a banking supervisor’s perspective.

First, it will not be possible to draw up a straightforward supervisory blueprint for dealing with future crises. As a general rule, every crisis tends to be different and there is no “one size fits all” approach to resolving them. We observed this recently in Italy and Spain. The institutions there differ in terms of the reasons why they ran into difficulties, their positions in the banking market and their standing in their respective economies. However, we must nevertheless learn specific lessons from each individual case.

Second, as this issue concerns appropriate, tailored solutions – and therefore discretionary scope as well – efficient decision-making processes are required. In this context, coordination between all of the institutional stakeholders plays a very important role. As crisis situations will always be unique, we need to ensure that all decision-making factors are incorporated sensibly and objectively.

Third, we need to grasp that banking supervision certainly can, and must, deliver far more than dealing with acute difficulties in the banking sector. As a general rule, it kicks in far earlier by ensuring compliance with the rules, thereby tackling unwelcome developments in good time.

European supervisors have set new standards as far as the last point is concerned. Capital add-ons, for instance, which are set following the annual supervisory review process, have become far more important than they once were. And that is as it should be.

Having learned from the crisis years, banking supervisors have strengthened their early warning systems and early intervention capabilities. Supervisors now go beyond the scope of an institution’s solvency and liquidity – these days, they normally also examine business models and governance, as unwelcome developments in these areas could have an impact on earnings and resilience in the medium term. One SSM project I could mention here explores the structural reasons for a disproportionately large risk appetite, which some institutions were shown to have in the past. This certainly does not mean that we are looking to intervene in an institution’s business activities; instead, it is our aim to identify at an early stage and mitigate to the greatest extent possible any developments that could pose a serious threat to a bank’s livelihood.

Imbalances in the banking sector, then, are an important testing ground for European banking supervisors - yet everyday supervisory measures are equally as important, even if their success cannot always be put into numbers as easily.

You can see this, for example, in the case of home country bias, a phenomenon which occurs when national supervisors unintentionally allow their national interests to influence their work. And it is precisely this trend that the SSM is designed to counter. It is not so much a case of supervisors breaking the rules, but more a question of how they interpret them, yet it is not always possible to identify a home country bias in each individual case. The new supervisory architecture is designed to counter this home country bias trend that national supervisors exhibit. Supervisory teams consisting of staff members from several different countries and, of course, the key role of the ECB have forged a more neutral approach and a more level supervisory playing field. I consider this a very welcome development.

Dear students, I hope that my remarks, which touched on only a small part of the work of banking supervisors in the new European era, have also managed to drum up some interest in our field of work. The tasks of national and European banking supervisors not only entail a high level of responsibility, but are also very varied and challenging. I do have my own particular reasons for saying this, however, not least as it is the specialists from the individual member states who keep the European supervisory mechanism up and running. Germany and, in particular, the Bundesbank contribute over 28 % of the total headcount of supervisors in the SSM. We therefore rate well qualified young professionals like yourselves very highly.

3 Resolution – what we have learned so far

Let us now take a look at the topic of resolution in the narrower sense. We all have unpleasant memories of institutions which unfortunately only appeared to obey the laws of a market economy and our regulatory framework when times were good. The new European Single Resolution Mechanism, or SRM, which consists of new rules, a new institution and a new resolution fund, was therefore established in 2015. How well is it functioning? What have we learned in the first few years?

There is one thing that I would like to stress from the outset which is all too easily forgotten when assessing the recent crises: neither in the case of the Spanish bank nor the Italian banks have we seen lasting spillover effects or negative repercussions in the markets. Branches opened their doors again on the next working day following the resolution – admittedly under a new name but otherwise it was business as usual. The fact that it all went largely unnoticed is the most remarkable thing about it, and that should by no means be taken for granted.

I would like to look a little closer at the resolution of Banco Popular Español in particular, which was the first financial institution to be resolved under the SRM. This was a concrete case in which the SRM achieved what some experts had even considered unimaginable up until then: that a decision could be reached within just a matter of hours on the fate of a multi-billion-euro financial institution with a countless number of branches - and that without any negative repercussions for the European or the Spanish financial system. Given the rapid pace at which the liquidity situation deteriorated, the institution even had to be resolved during the week between two normal working days. In this specific case, the fact that a purchase offer had already been made meant that it was possible to find a solution to the crisis with comparatively little intervention. The management board and supervisory board of Banco Popular also acted in a responsible manner.

The case of the two Venetian banks was at least just as challenging. I am pleased that the market exit of these two banks ultimately turned out to be a success. The precarious banks that were causing us concern in the past no longer exist in their previous form. Unlike in the case of Banco Popular, these banks were not resolved in accordance with European rules, but were liquidated under Italy’s national insolvency regime.

Here, it was possible to wind up the institutions using government funds as the European resolution rules apply only to banks that are classed as being systemically important. The switch from a European regime in crisis cases to a national regulatory framework consequently created a whole new ball game.

One requirement, nonetheless, was that the European state aid rules be adhered to. Consequently, the banks’ owners and their subordinated creditors, too, had to be bailed in. The fact that substantial amounts of taxpayers’ money were drawn on at the same time was, of course, justified from a legal perspective. However, the declared objective of the European resolution rules was to keep the taxpayers’ contribution in bankruptcy cases to a minimum. In order to ensure that national taxpayers are adequately protected, it is therefore essential that national insolvency rules be aligned with the European resolution rules as a matter of urgency. 

4 The current banking union environment – the most important details 

Ladies and gentlemen, the cited examples show that many things are working well in the banking union, but there is still a lot to be done. In particular, right-sizing the bail-in of investors and other creditors under national insolvency proceedings is a process which has by no means been stretched to its limits. This, however, is important if banks are to live up to their claim of being fully-fledged members of our market economy. What this also means is that losses should not be socialised, but rather that investors and creditors should personally be held responsible for the risks arising from their decisions. I am convinced that we must press ahead along the path that we are now on.

A debate on the banking union needs to address many details. This also includes taking a look at the bigger picture. If we are concerned with the resilience and credibility of the new regime as a whole, we must always be aware that the banking union was designed for a banking system scarred by the crisis. We committed the European banking sector to “self-healing” from the very outset. To this end, the balance sheets of the large euro-area institutions were thoroughly examined prior to the launch of the new oversight body. I recall valuation adjustments of €48 billion and mandatory capital increases totalling €25 billion in the period from November 2014 to June 2015.

However, we were, of course, unable to foresee new burdens resulting from factors such as persistently weak economic activity and knock-on effects of the crisis further down the line. Non-performing loans are still saddling banks in a number of member states. The level of non-performing loans for EU banks currently stands at around €900 billion. These burdens were a notable feature of the four most recent crises. Ultimately, self-healing does not happen overnight, but is often a long and arduous process.

This is also true of the resolvability of credit institutions. Although the new rules were quickly reflected in price effects for subordinated bonds following their entry into force – which is an indication that they had the desired steering effect from the very outset –, the transition into the new world is by no means completed. The Single Resolution Fund, for example, is still in the development phase. And Europe’s credit institutions are still building up their MREL and TLAC holdings – in other words capital that is to be made available for bail-ins. The corresponding MREL requirements are expected to be determined by the resolution authorities by the end of this year.

5 Conclusion

Let me come back to my initial question: is the European banking union working?

The takeaway from the recent crises is ambivalent. For sure, we can breathe a sigh of relief that we were able to overcome four bank failures without any far-reaching consequences and that ailing banks exited the market successfully. At the same time, we have to draw conclusions from the recent crises in order to further strengthen the banking union.

In my opinion, closing the gap between national insolvency rules and European resolution rules is a major issue that should be addressed as a matter of urgency. As sensible as it may seem to deal with crises differently depending on the specific circumstances, certain basic principles do need to be adhered to in a consistent manner. These include, for example, bailing in investors and creditors as envisaged in the European resolution framework.  

In the discussion about the functional viability of the banking union, I would, however, like to warn against making overly one-sided and generalised judgments. We can only strengthen the large-scale “banking union” project if we are willing to devote ourselves to the specific problems and not let up until they have been fixed for good. This means taking account of the individual circumstances - both for the four cases discussed earlier and the over 3,000 other credit institutions in the euro area. It also means accepting that the banking union cannot pull simple solutions out of a hat when the going gets tough. The high level of non-performing loans in Europe bears witness to the major challenges that still need to be overcome.

Thank you for your attention. I now look forward to our discussion on this topic.