Andreas Dombret: The other side of the coin - why European supervision needs international regulation

Speech by Dr Andreas Dombret, Member of the Executive Board of the Deutsche Bundesbank, at the Annual Conference on the Banking Union, organised by Freshfields Bruckhaus Deringer LLP and the Institute for Law and Finance (ILF), Frankfurt am Main, 15 May 2017.

1. Introduction

Dear Danièle Nouy, dear Adam Farkas,

Ladies and gentlemen

Do you feel uncomfortable whenever a speaker spends most of his time off-topic? Then please be forewarned - this just might be one of those speeches. I am pleased to have been invited today to talk about the current state of the Single Supervisory Mechanism, the SSM. But I want to broaden the topic to some extent and speak about the other side of the coin - which is regulation.

But let's take a step back: The banking union was established as a consequence of the European debt crises. We wanted banking supervision in the euro area to achieve greater transparency, neutrality and fitness. Amongst other objectives, the banking union was designed to counter the close ties between member states and their banks - which proved to be one of the main problems of the European debt crisis.

But a European supervisor is not a universal remedy - it is not the only way to cope with these issues. The other side of the coin is regulation. And in fact, if we do not fill the existing regulatory gaps, the effectiveness of the SSM might very well be weakened. This is because there are various channels through which the national state-bank relationship thrives.

That's why I believe it is important to address regulation as the other side of the coin. If we want an effective prudential architecture for the European financial sector, we will need to address the regulatory gaps that still persist.

In my speech this morning, let me begin by reviewing the functioning of the SSM and discussing its challenges. Then I will focus on the regulatory treatment of sovereign exposures as one example of a regulatory gap that still needs to be addressed.

2. Still being minted: European supervision

I remember well being a participant at the ILF and Freshfields conference in May 2015, when the Single Supervisory Mechanism was still in its infancy. Back then, we were mainly concerned with the workings of the new system.

Today, the focus has shifted. Structural challenges abound in the European banking sector - such as the high volumes of non-performing loans, low profitability, and IT and cyber security. For European supervisors, preserving the "tough but fair" character is to a large extent a matter of day-to-day supervisory practice. Danièle will surely stress this point when she addresses you later.

And there are external events such as Brexit which demand our attention. I consider Brexit to be manageable for banks, but it requires timely, thorough and comprehensive preparations. And because of the uncertainty it involves, supervisors have their own part to play in ensuring a smooth transition to a post-Brexit era. It requires us to be responsive and pragmatic as well. Therefore, I strongly welcome the commitment shown by SSM supervisors to cooperation and open exchange.

There is no need for me to go into detail on these subjects. As I mentioned just now, these are topics I very much imagine Danièle will touch upon.

Instead, I wish to review the set-up of the European Single Supervisory Mechanism and offer some ideas as to how it can be developed further. But for that purpose, I find it helpful to remind ourselves of the organisational principle of the SSM. It was not intended to be a single, hierarchical organisation, but rather a supervisory network. There are still many misperceptions on that matter.

Without doubt, the ECB has key competences regarding the SSM. The ECB assumes responsibility for the overall functioning of the project. And apart from directly supervising the large, "significant" credit institutions, it has also taken over various powers from the national competent authorities, the NCAs - one such important power being licensing.

But that does not make it a hierarchical undertaking. Instead, different modes of cooperation were installed. For example, joint supervisory teams include personnel from NCAs that now receive their instructions from the ECB, while inspectors from NCAs continue to be directed by their employer. Other NCA staff again are on secondment at the ECB.

The different modes of cooperation also become visible with respect to competences. NCAs still have major tasks and competences beyond directly supervising the 3,200 smaller, "less significant" institutions in the euro area. Think of the roughly 1,500 written procedures from last year: every single decision is approved by the Supervisory Board, which includes members from 19 NCAs, thus boosting the transfer of knowledge from NCAs to the ECB and ensuring that decisions are subject to multiple levels of protection, not just dual control.

Also, the NCAs are the source of the vast majority of supervisory staff. For example, the Bundesbank has staff members in 33 out of 127 joint supervisory teams. And NCAs contribute the vast majority of staffing in the SSM's overall workforce. This means that national authorities still - and for good reason - play an essential role in the operational functioning of the SSM.

Today, and given the huge number of challenges I mentioned earlier, the SSM is very much reliant on the full and vigorous support of the national competent authorities. NCAs thus need to adhere to their staffing commitments.

The network character of the SSM also demands fostering a common supervisory culture. We should strengthen our aim of promoting a common supervisory culture, which we have labelled a "one-team" approach. For that purpose, we should remove the boundaries between supervisors in their daily routines and interactions whenever we see them.

Another issue we still need to work on in this respect is language. Among the elements of cultural diversity in Europe, language is among the most deeply-rooted. Here at this conference, we do manage to communicate in a common language, even though there are speakers from several nationalities. But when it comes to the legal details of banking regulation and supervision, the meaning of words becomes crucial. In fact, even the points you are listening to right now have been checked linguistically by our language specialists at the Bundesbank. It is therefore understandable that bank representatives will want to stick to their home language when they interact with authorities.

And even if top managers feel confident enough to engage with our supervisors in English, our inspections regularly look in-depth at departments where the home language prevails, for example when we look at documentation. This gives domestic inspectors a natural advantage in their jobs. If we want to further reduce potential instances of "home bias", we therefore need to target those language barriers for foreign inspectors and supervisors. For the SSM, this means we need to strengthen our efforts to boost the language skills of our supervisory staff.

Having spoken about challenges both outside and inside the SSM, let me now turn to the more general topic of my speech. How has the SSM performed overall so far? I guess it is fair to say that supervision has benefited in many ways: Supervisory quality has improved, cooperation has been extended, and the SSM has contributed positively to financial stability. Constrained by the given regulatory environment, the ECB has been persistently undertaking measures to address various structural issues. For example, it has originated guidelines for institutions on dealing with NPLs. Also, it had launched a project targeting options and discretions in European banking regulation that may well be a source of national bias.

Thus, the SSM has already made important contributions to enhancing financial stability and to reducing "home bias".

But reliable supervision is not the only road that leads to a sustainable financial order in the euro area. Just remember the multiple ties binding member states and their national credit institutions together. On the one hand, a sovereign solvency crisis risks triggering a solvency crisis at banks. On the other hand, struggling financial institutions were able to force governments to bail them out, thus endangering state solvency. This doom loop - an ever-worsening nexus between banks, governments and the entire national economy - showed up quite clearly during the European debt crisis. Today, strong ties between banks and states continue to exist. A survey conducted by the EBA in 2016 found that the share of home sovereign exposures is disproportionately high. In some member states, it is very close to 100 per cent. Germany is no exception here: The sample of German banks revealed that 74 per cent of their sovereign exposures was to German general government.

Just as with the monetary policy of the Eurosystem, the SSM and the banking union may not iron out all the deficiencies and negative impact channels surrounding the ties between banks and states. Supervisory authorities need to reject the role of a universal remedy. Banking supervision can only be as effective as the reach of the rules. If regulatory gaps still exist in the euro area, we will only be able to achieve a well-balanced political and financial order if we plug those gaps.

3. The other side of the coin: Adequate regulation

Let me demonstrate this reasoning with respect to the regulatory treatment of sovereign bonds, which plays a key role in strengthening the overall architecture.

There is not much dissent, in fact, over the question as to whether sovereign debt is subject to economic risk. Puerto Rico declared bankruptcy just a few days ago. In Europe we have been witnessing debt haircuts and other symptoms of sovereign overburden. By and large, the risk of sovereign default may be low, and sometimes very low, but it is not zero.

If we want regulation to be risk-adequate, economic risks will need to find their way into the rules for banking. Our existing regulatory set-up is quite well-developed in this respect. The standard requirements for credit risk already comprise eight risk weight categories. The specific risk weights applied to each credit category were substantiated by historical evidence on credit risks. Yet historical evidence is not taken into account where sovereign bonds are concerned. Euro-area credit institutions do not need to hold any capital against the default risk of their member states. This is remarkable, since sovereign bonds account for a significant part of bank assets in the euro area. 

For sure, the leverage ratio - which was introduced as part of Basel III - may, in principle, set a ceiling for sovereign bond holdings at credit institutions once it becomes legally binding. This is because the leverage ratio caps indebtedness at a certain level. A bank which barely meets this limit will not be able to buy any further sovereign bonds without first acquiring additional capital. But the leverage ratio is designed only to be a backstop for the minimum capitalisation at banks. Thus, it is not an instrument that directly addresses sovereign risk.

This regulatory gap has numerous consequences. Most prominently, it creates inadequate incentives for banks and feeds into the doom loop I mentioned earlier on.

So what about our new supervisory regime in the euro area? Couldn't our supervisors counteract this aspect of the unhealthy relationship between member states and their national banks, given that the SSM is intended to be more independent and less negligent regarding the close ties between banks and states? The uncomfortable answer is: You cannot expect European supervisors to fill a regulatory gap like this in a fully satisfactory manner.

In theory, supervisors may target sovereign risks as part of their supervisory actions - the buzzword here is pillar 2. Indeed, when institutions are asked to calculate their idiosyncratic risks, these also include their sovereign exposure. This risk calculation feeds into the supervisory evaluation process, which may result in additional capital requirements. But of course, sovereign exposure is only one element of what turns out to be a complex evaluation process for every supervised entity. Heterogeneous treatment of this risk category at different institutions across the euro zone is a phenomenon that cannot be ruled out. Also, interactive risk assessment is most beneficial only for those risks that are indeed idiosyncratic. By contrast, sovereign risk is predestined for standardised treatment.

The damage caused by inadequate rules is not confined to banks. Artificial additional demand for sovereign debt may lead to distorted price signals and, as a consequence, to private investment being crowded out. And of course, it gives governments the wrong incentives. Even central banks are affected, because they accept sovereign bonds as collateral when lending money to banks.

Thus, neglecting sovereign risk is not only inconsistent both from an economic and a regulatory view; it also produces sizeable side-effects in other domains. The European debt crisis is the best evidence there is on the detrimental linkages between states and domestic banks. We can only contain these unhealthy relationships if we target the root causes, such as the lack of regulatory treatment of sovereign risk.

What could be the therapy? It would consist of two components: Non-zero risk weights as well as concentration limits for sovereign exposures. Concentration limits directly relate to a fundamental principle of banking: Never put too many eggs in one basket. In the status quo, euro-area sovereign bonds are exempt not only from non-zero risk weighting, but also from the existing concentration limits.

4. The road ahead

Of course, simply believing that the euro-area architecture contains structural flaws will not, in itself, invoke change. Transitioning towards a more desirable order may get tough for those affected.

Banks do not only make some of their money with sovereign bonds; they also use this paper as collateral in interbank markets and for their borrowing from central banks. Also, increased holdings in sovereign bonds have helped banks comply with stricter capital requirements and newly enacted liquidity requirements. For example, German banks were able to improve their capital ratios partly by increasing their holdings of euro-area sovereign bonds relative to investments, which lowered their RWAs. Amending the regulation of sovereign risk gives rise to a number of tasks in the management of banks.

A similar storyline can be told about states that are highly indebted. Putting an end to subsidised lending will often be onerous. Some may also be worried about cliff effects.

We need to take those temporary burdens into account. Whatever path the European financial architecture is taking, we need to ensure that it is not only viable, but also feasible in the first place.

But the fact that resistance to reforms happens for a reason does not mean that it is a superior reason. This is evident in the field of government financing. Markets have not been able to discipline governments. But who else, if not creditors, would be more suited to pushing governments to take a sustainable financial path? Why should we expect states to leave the well-trodden path of least resistance - which is expanding debt instead of committing to painful reforms - when there are no regulatory restrictions such as a non-zero capital requirement for sovereign bonds in place that provide actual incentives to do so?

So whatever transitional challenges there may be, they don't make the idea of a more risk-adequate treatment of sovereign bonds less of a desirable ultimate aim. Also, framing the reform as a mere distribution battle between member states is deceptive. In the long run, everyone stands to benefit from a sustainable order.

Instead, we need to focus on change management. How do we smooth the transition toward a sustainable financial order? In my view, just as nobody should want a drug addict to die from going "cold turkey", we should take into account the challenges facing those who will be worst affected by a change in regulation - and offer demanding, but feasible transition schedules.

Of course, regulation of sovereign risk is also a global issue. Any framework of rules for the financial sector these days needs to be comprehensive and consistent. There is a straightforward and sobering reason for this: You can count on the global financial industry to quickly locate regulatory gaps. The Basel Committee has put sovereign risk on its agenda for 2017 and 2018. I strongly welcome this step, and call upon all parties to work together to find a good solution.

For the euro area, reforming the regulation of sovereign risk will be even more vital. This is because a currency union is particularly vulnerable to national distress in case that market forces break down. A common currency, for example, does not allow currency depreciation as an automated correction mechanism. Therefore, achieving reform towards better self-governance in the banking sector, markets and state finance will be an indispensable step towards a more sustainable euro area.

5. Conclusion

Ladies and gentlemen

I have listed several organisational challenges which the SSM will need to come to terms with, if it is to live up to its character as a network.

  1. National authorities need to honour their staffing commitments, because these are what drives the operational functioning of the SSM.
  2. We need to remove boundaries between supervisors in their daily routines with a view to nurturing a "one-team" culture.
  3. We need to enhance our language skills to account for the cultural diversity that is part and parcel of life in Europe.

But the SSM - whether it runs smoothly or is still in need of improvement - is not a universal remedy for home bias and the nexus between banks and their home sovereigns in the euro area. Only if they are flanked by a coherent set of rules for the banking sector will the SSM and the banking union as a whole be in a position to perform to our fullest satisfaction.

Putting an end to the privileges afforded to sovereign risk at banks will be a milestone. It is one thing to agree on all the components of a consistent regulatory framework - how quickly that aim can be achieved in the face of political pressure, is another. Let us find answers to both these questions.

Thank you for your attention.