Sabine Lautenschläger: European banking supervision - achievements, challenges and the way forward

Speech by Ms Sabine Lautenschläger, Member of the Executive Board of the European Central Bank and Vice-Chair of the Supervisory Board of the European Central Bank, at the ESE Conference 2017, Vienna, 28 September 2017.

The views expressed in this speech are those of the speaker and not the view of the BIS.

"What does it mean to be human?" This question has preoccupied philosophers for hundreds of years. One of the more popular answers is that humans are rational animals - a view that goes back to Aristotle.

Let us take that question and adapt it to our world: what does it mean to be a European banking supervisor? The answer is certainly not that banking supervisors are rational bankers.

But apart from that, being a European banking supervisor means many things. First of all, it means looking across borders. It means being tough and fair, unbiased by national interests. For us, it means being part of a team of supervisors from 19 countries and 27 institutions. It means working in the public interest - and having an exciting job.

But it also means dealing with many challenges at the same time. And it is these challenges I'd like to talk about in my speech. But as there are many and as our time is limited, I have to make a choice. So let me first tell you what I won't do today.

I won't discuss the bringing together of cultures and languages that are very different. I won't discuss dealing with banks that struggle to remain profitable. I won't discuss digitalisation, and I won't discuss Brexit.

Instead, I will focus on three issues: first, the rulebook for banks and how it relates to supervision; second, the toolbox that supervisors have at their disposal; and third, the role of the market.

Supervisors and the rulebook

A solid rulebook is a cornerstone of a stable banking sector. It ensures that banks acknowledge the risks they take and build up adequate buffers. A solid rulebook lays the foundations for banking supervisors to do their job.

Since the financial crisis, the rulebook for banks has been thoroughly revised. Today, banks face much tougher rules than before. They need to hold higher capital buffers that consist of better capital. At the same time, they need to hold liquidity buffers - for the first time ever. And these are just two of the things that help to make banks more resilient.

So the regulatory base has been made more solid. But it needs to be more than solid; it also needs to be level. The rules for banks should, in general, be harmonised across countries. Only then would banks be able to compete on a level playing field.

But what's the scope for harmonisation? Here, we should be bold, in my view. The financial sector is global in scope, and so should be the rules that govern it. And after the crisis, it was precisely this idea that drove the work on a global rulebook: Basel III.

But since then, the political landscape has changed. It seems that there are some who have stopped subscribing to the idea of a global approach to regulation. National interests might get the upper hand, and that's not good. Against that backdrop, we need to finalise Basel III as quickly as possible.

But having a global set of standards is just the first step. The second step is for all countries to implement it in a consistent manner. That includes the euro area. Here, a level playing field is even more important, given that we have a single currency and a single market.

And there is indeed a single European rulebook. However, when you take a closer look, it turns out that the single rulebook is not so single, after all. So, in fact, we still have a kind of regulatory hotchpotch in Europe. How can that be?

Well, one reason is that parts of the single rulebook come in the form of EU Directives. And these directives still have to be transposed into national law. That in turn opens the door for significant differences between countries.

One of the many examples relates to what is known as "fit and proper" requirements. Bank directors have to fulfil certain criteria of competence and probity before they can take up their job. However, these criteria differ across the euro area. In some countries, they are quite strict, in others less so. And as the ECB is bound by national law, we cannot provide harmonised supervision in this case.

But the patchwork consists of more than that. The rulebook also contains what are known as options and discretions, O&Ds for short. These O&Ds provide governments and supervisors with some leeway in applying the rules. Again, this has led to differences across countries.

Here we have made some progress, though. Together with the national supervisors, we have agreed to exercise a number of options and discretions in a harmonised manner. But there are still O&Ds that lie in the hands of governments. So it is up to them to bring about more harmonisation. And I admit that I am quite disappointed that so little has been done in this area.

And last but not least, there is a third item that helps to explain Europe's regulatory mix. We still see that some countries are issuing national laws that affect the core of European banking supervision. There are, for instance, national laws that govern risk management. And these laws apply to banks that are directly supervised by the ECB, of course. Now imagine that there is a banking group with entities in Germany and Spain. This banking group would be subject to different rules when it comes to managing credit risk or to liquidity reporting. Needless to say, this seems ill-suited to a banking union and a level playing field.

Now what does this regulatory hotchpotch mean for European banking supervisors? Well, it makes them less effective and less efficient. In the worst case, they have to deal with 19 national rules instead of a single European one. This makes it hard to treat banks equally. It leads to more bureaucracy and higher costs. It opens the door to regulatory arbitrage, and it distorts competition.

My advice is therefore clear: we must further harmonise the rulebook. Instead of EU Directives, we should rely more on EU Regulations, which can be directly applied in all Member States. We should deal with the remaining O&Ds, and we should stop passing more and more national laws, especially those which apply to banks that are supervised at European level.

Instead, we should strive to cut down on some of these national laws, such as those concerning the reporting requirements that have become redundant as a result of the European Reporting Framework. Scrapping such laws would increase efficiency and reduce costs for both banks and supervisors.

So supervisors need a strong and level regulatory base. However, we must be careful that it does not turn into a cage. How could that happen? How could the rulebook evolve in a way that would confine supervisors?

The problem here is "too much detail", a kind of regulatory "overload" - involving too many stakeholders and too much bureaucracy. There is a tendency to add more and more details to the rulebook. This is driven by a desire to have rules that cover every eventuality. Whatever happens, there should be a rule to deal with it. It would trigger supervisory action. And if it didn't exist, well, then there's no trigger and no action.

However, trying to have a rule for every eventuality is an elusive goal, of course. The business of banking is so complex and fast-changing that no rulebook would be able to fully cover it. The unexpected will always happen. If not by chance, then because banks make it happen on purpose.

And precisely that is one of the problems entailed by an overly detailed rulebook: regulatory arbitrage. The more detailed the rules, the more ways banks can game them. This creates new risks, which are then not covered by the rules.

On top of that, an overly detailed rulebook would make it much more difficult for supervisors to react decisively and quickly to a changing environment. Supervising a bank could turn into a mere box-ticking exercise. And supervisors would turn into experts for tiny details; they could end up focusing on the minutiae and forgetting the big picture.

And that would be a waste. It would be a waste of all the experience, expertise and insights that supervisors have. Why not strengthen their judgement? Why not keep the flexibility they need to react to newly emerging risks in an ever-changing banking sector? And do all that within a sensible legal framework, of course.

At the end of the day, we need both. We need clear-cut rules, and we need sound supervisory judgement. Adding more and more details to the rules might come at the expense of supervisory judgement. With the recent reforms we have struck a balance in this regard. If we were to upset this balance, it would become much harder to ensure that banks remain safe and sound.

Supervisors and their toolbox

But European banking supervisors need more than an adequate rulebook to do their job. They also need the right tools, of course. They need a European toolbox.

And here, a lot has been done. We have harmonised the main tool of banking supervision: the supervisory review and evaluation process, SREP for short. In the SREP we take a very close look at each bank. We analyse its business model, its governance, its risk management as well as risks to its capital and liquidity positions.

The SREP provides a common yardstick to compare banks across the euro area. It helps us to identify best practices and spot common problems. By referring to the SREP, we develop specific supervisory requirements and measures for each bank. And by the way, supervisory judgement plays a big role in the SREP.

But the SREP is a massive tool, and we still need to refine it. Let me give you an example. As part of the SREP, we review the processes that banks use to assess and maintain adequate capital and liquidity buffers. These processes are known as ICAAP and ILAAP¹. Based on what we learnt from our SREP exercises, we will refine our expectations regarding ICAAP and ILAAP, and our methods for reviewing them.

And we also need to consider how best to assess the risk management of banks in the SREP. We need to do more than just define what we expect with regard to the management of IT risks, outsourcing or leveraged finance. We need to define fully fledged minimum expectations for the risk management of banks.

But our toolbox comprises more than the SREP. So there are other things that we need to improve. First, there are tools that are applied in different ways in different countries. On-site inspections at banks are one example. In some countries, the preparation takes some time, while the actual inspection is quite short. In other countries, it is the other way round. This makes it difficult to coordinate and to create international teams of on-site inspectors.

Second, there are tools which exist in some countries but not in others. In this regard, the moratorium tool comes first to mind. It allows supervisors to suspend all the activities of a failing bank for a short period of time. It helps to handle such failures in a timely and orderly manner. Therefore, moratorium powers should be part of the European toolbox.

Another tool that we need to have is called "deductions from own funds". It would help to ensure that banks make adequate provision for risks - adequate from a prudential point of view. The national accounting frameworks allow for some flexibility with regard to provisioning. And what a bank then chooses may not be sufficient from the supervisors' viewpoint, taking into account the bank's risk profile. That's why supervisors should be able to impose prudential deductions from own funds.

And third, there are some tools which are part of the European toolbox not once but twice. And this can cause problems. A relevant case relates to banks that get into trouble and the question of how to intervene as early as possible. As a rule of thumb, banks don't get into trouble out of the blue. Rather, it is a gradual downfall in most cases. Thus, we have tools at our disposal that allow us to step in early and send warning signals to the banks.

This is known as early intervention. However, the relevant tools are the same as some of our standard tools. And this overlap might prevent us from using early intervention tools. The reason is that we are bound to use the least intrusive tool, which is always the standard one. This is particularly relevant should the market authorities demand that the bank disclose whether early intervention tools have been deployed. So, the overlap between early intervention tools and standard tools should be removed.

To sum up, we still need to work on the toolbox of European banking supervisors. We have to expand, harmonise and streamline it.

Supervisors and the market

Now let us come back to my initial question. Being a European banking supervisor means doing an important job. Equipped with an adequate rulebook and the necessary tools, supervisors help to make banks safe and sound. Such banks in turn form the foundations of a strong economy. But we should not become too self-centred.

Banks are private enterprises, and the banking sector is a market, after all. As such, it deploys forces that can help to keep risks in check and secure stability. There were even times when many believed that market forces alone could bring about a stable and efficient banking sector, ensure sustainable business models and make banks resilient to the ups and downs of the economy. That belief was shattered by the crisis.

But don't get me wrong. I believe in the market; in theory, the forces of the market work fine. The problem is just that, in practice, they do not always work so well, for several reasons.

First of all, the market is made by people interacting, and people have their limits. Ample research shows that human brains are not very good at handling risk. We tend to over- or underestimate it; we tend to get carried away by potential gains; and we tend to overreact when things change.

And then, people's behaviour depends on incentives which are sometimes distorted. The most prominent example is this: whenever you can keep the gains but pass potential losses on to someone else, you might be tempted to take on too much risk.

So, in practice, the market has its shortcomings. But is that a reason to do away with it? No, not at all. We should try to make it work. For the banking sector, this means aligning incentives.

And as I just mentioned, the best incentive to keep risks in check is potential failure and financial loss. As the economist Allan Meltzer noted: "Capitalism without failure is like religion without sin. It doesn't work." But in banking, there was a lack of failure.

During the crisis, many banks that should have failed did not fail. Instead, governments propped them up with public funds - at huge cost to taxpayers. They did so out of fear of a systemic crisis. They were worried that when a large bank fails, it would drag others down and destroy the entire system.

So there was a situation where banks could pass on any losses to someone else while keeping any profits. They had an incentive to gamble with other people's money. So, we need to make it possible for banks to fail without causing the whole system to collapse. And we need to make sure that profit-makers are also loss-takers. That would align incentives, strengthen market discipline and keep risk-taking in check.

In Europe, we now have a new system for exactly that purpose: the Single Resolution Mechanism, SRM for short. The SRM provides the tools to resolve banks in an orderly manner. Earlier this year, it passed its first test when a couple of banks failed. This was a huge step towards imposing market discipline on banks and making them safer and sounder.

And I can confirm that dealing with a failed bank is among the most complex things you can do as a supervisor. The first step is that the ECB declares the bank failing or likely to fail. And that, in itself, is a delicate matter.

A bank can fail for a million reasons. Sometimes it is a slow death that drags on over weeks and weeks. Sometimes it all happens within a couple of days. In the end, it is up to the ECB to decide whether the moment has come, and that is a decision which cannot be taken lightly.

Taking it too early might infringe the rights of investors and creditors. Taking it too late might lead to a systemic crisis; it might harm the economy and society. So all in all it is a decision that requires supervisors to carefully judge the case at hand.

Once the ECB has declared a bank failing or likely to fail, another body takes over: the Single Resolution Board, SRB for short. It is the SRB that decides how to deal with the bank.

In a nutshell, this is how we handle failing banks in Europe. Once you start looking at the details, it becomes much more complex, of course. It requires all parties involved to closely cooperate. And it requires all parties, including the banks, to be well prepared. That's why it is so important for banks to carefully draft their recovery and resolution plans.

But as I said, recent experience has shown that our approach works. The ECB, the SRB, the European Commission and the relevant national authorities proved that they can cooperate closely. They have demonstrated that they can deal with failing banks and handle the whole process smoothly.

That said, things can and should be improved, of course. And I am not just thinking of the actual failure of a bank. I am also thinking of what can be done before it comes to that.

And here, we need to think about precautionary recapitalisation, for instance. Under very strict conditions, the European rules on state aid allow banks to be given public funds. However, these aid rules might need an update that would align them with the new European resolution regime.

In this context, we also need to reflect again on how we define solvency. After all, banks need to be solvent in order to receive precautionary recapitalisation. Here, we need to reach a common understanding of how to define solvency, in particular its forward-looking aspect.

As for the banks, they need to reflect on how to handle their liquidity during a crisis. Most importantly, they need to make sure that they have a backstop in place that is readily accessible.

And then, when push comes to shove, we need to improve the tools that are needed to actually resolve a bank. I already mentioned that some tools still need to be added to the European toolbox. The moratorium is one of them.

And finally, we must keep in mind that only systemically relevant banks will be resolved at European level. All other banks will be subject to national insolvency regimes. It might thus be warranted to harmonise these regimes across Europe to ensure a level playing field.

All the things I just mentioned will have an impact on how supervisors assess and react to the risks of banks and the risks of bank failures.

Conclusion

Ladies and gentlemen,

What does it mean to be a European banking supervisor? That was one of my questions at the outset. And the answer was that it means many things. It is a job that is challenging and rewarding at the same time.

Today I have mostly talked about the challenges. I have talked about the need to write a solid and harmonised rulebook. I have talked about the need to expand, harmonise and streamline the supervisory toolbox. And I have talked about how to make the market work again.

To end on a more positive note, let us turn to the rewarding aspects of being a European banking supervisor. Above all, it means being part of a united Europe. It means working towards a stable and truly European banking sector - a banking sector that spans 19 countries and serves more than 300 million people.

And all these people rely on banks. They rely on banks to invest their savings, to help them start a business or buy a home. It is the job of a European banking supervisor to contribute to a stable banking system that can reliably and efficiently serve these needs.

Thank you for your attention.

---

¹ ICAAP: Internal capital adequacy assessment process; ILAAP: Internal liquidity adequacy assessment process.