Regulation, supervision and market discipline – striking a balance
Statement by Sabine Lautenschläger, Member of the Executive Board of the ECB and Vice-Chair of the Supervisory Board of the ECB, at a conference hosted by the Financial Stability Institute, 18 September 2017, Basel
Almost nine years ago to the day, the investment bank Lehman Brothers failed.
What followed was a financial crisis that wreaked havoc on economies around the globe. But it did more than that.
The crisis also shattered many beliefs about how the banking system works and how it should be regulated and supervised.
There was a time when many people believed that the market could bring about a stable and efficient banking sector on its own. They thought that market forces would ensure banks had sustainable business models that were resilient to the ups and downs of the economy. We now know that this did not turn out so well.
In fact, there are three pillars that support a stable banking sector: regulation, supervision and market discipline. The crisis made clear that all of these pillars were in need of some repairs.
The good news is that we have come a long way – today, the banking sector rests on more solid foundations than before. But some issues still need to be resolved. Let us briefly look over them, beginning with regulation.
Once the crisis had started, everyone agreed that the rulebook needed to be rewritten. And back then, everyone agreed that the revised rulebook needed to be a global one. The crisis clearly demonstrated that banking is a global business, so the rules that govern it should also be global.
Today, however, it is much less clear whether everyone still subscribes to the idea of a global rulebook.
And this worries me.
It worries me because it shows that people are starting to forget the lessons we learned from the crisis.
That’s why we need to finalise the global rulebook, Basel III, as quickly as possible.
But that is just the first step.
The second step is to implement and apply the rules in a broadly consistent manner at regional and national level. And this is a challenge – sometimes we see quite a different interpretation of the standards when it comes to implementing and applying them at national level.
You won’t be surprised that I will focus on the euro area. The euro area countries have become closely intertwined. For many years now they have shared a single market and a single currency. Since the crisis, they have even begun to build a banking union, which includes a single supervisor.
But when it comes to regulation, we still see differences between countries.
It is true, of course, that we have a single rulebook in Europe. As one might expect for a union, we had to come further than other countries.
However, parts of that rulebook come in the form of EU directives. And these still need to be transposed into national law, which explains some of the differences we see.
So regulation in the euro area remains fragmented. But this runs counter to the idea of “same business, same risk, same rules”; and it runs counter to the idea of the European banking union. So there is a clear case to be made for further harmonising the rules. We need a well-written rulebook to keep risks in the banking sector in check.
But rules are just one of the pillars that support a stable banking sector. The second pillar is supervision. And here we need to strike a delicate balance because these two pillars interact in a subtle way.
The general idea is that rules create the foundations for supervisors to do their job. There is a danger, though. If the rulebook becomes too detailed, it might wall supervisors in and limit their options too much, rather than provide them with solid foundations.
And this can happen fairly quickly. By way of example, just think of speed limits. Let’s say you want to set a speed limit for motorways. Does 120 kilometres per hour sound about right? At first sight, this is a very simple and straightforward rule.
But then you start to think – what if there’s a sharp and dangerous bend in the road? Wouldn’t that call for a lower speed limit? And what if the road becomes slippery when it rains? Perhaps the speed limit should depend on the weather? And shouldn’t it also depend on the kind of vehicle? One could easily think of many more questions like these.
So even a seemingly simple rule can quickly start to become more and more complicated.
The world is complex and uncertain, particularly when it comes to finance and banking. That leaves us with two options.
Either we come up with more rules to try to cover every potential situation.
Or we allow supervisors some discretion within a set framework and let them judge specific situations on their merits.
In the end, these two options need to be in balance, and we must be careful not to disturb this balance. And here is where I see some cause for concern.
Regulators might be tempted to lay down rules to cover every eventuality – think of the example I just gave you. And their motivation would certainly be laudable. But their goal would be unattainable.
It is just not possible to cater for every potential situation. There are too many possibilities, most of which we cannot anticipate. No one knows what financial innovations might emerge; no one knows where technological change will lead us; and no one knows how the economy will evolve over the coming decades.
If we try to cover everything, the rulebook will, at some point, become too detailed, too big and too complex. From that point onwards it will stop making the banking sector a safer place and instead create new problems.
First, overly complex rules facilitate regulatory arbitrage. Yes, we could add page after page to the rulebook, setting out detailed rules, definitions and thresholds governing how banks should deal with risks and specifying, for instance, which tools supervisors should use in certain situations, and when. But with each new page, more loopholes open up. It becomes easier for banks to game the system. This defeats the purpose of a rulebook and poses a threat to stability.
Second, there is the problem of interaction. The more rules there are, the more they can interact and interfere with each other. If there are two rules, there is just one way in which they can interact. If there are 1,000 rules, they can, in theory, interact in at least 500,000 ways. It becomes ever more difficult to foresee the overall effect of the rules, and the danger of unintended side-effects increases.
And third, there is the issue of costs. The thicker the rulebook, the more expensive it becomes for banks to comply with it and for supervisors to do their jobs. These costs are justifiable if each additional rule reduces the risk of another crisis. However, at a certain point the costs of the rules begin to outweigh the benefits.
So, instead of having overly detailed rules, we should have a strong basic set of rules that allows supervisors some discretion. .Supervisors have the relevant skills and experience, and in-depth knowledge of the banks they supervise. We should ensure that supervisors can continue to apply their judgement. They should be empowered to assess the risks that a bank poses to its stakeholders and the financial system. This will enable them to deal with risks at an early stage.
But if the rulebook becomes too extensive and too detailed, supervision turns into a simple box-ticking exercise.
A large, detailed and seemingly comprehensive set of rules would make supervisors less flexible and effective; it would leave no room for discretion. Supervisors would no longer be able to react quickly to newly emerging risks in an ever-changing banking sector. They would be walled in by rules that were devised to cover every possible situation but, in reality, were outdated right from the start.
So, looking ahead, we must fight the urge to make the rulebook ever more detailed.
But don’t get me wrong – we shouldn’t unwind any of the reforms. What has been done was necessary. But we should be cautious about going too far. A few clear principles can be more effective than 1,000 complex rules. The two pillars of regulation and supervision are equally important and we should not favour one at the other’s expense.
And this brings us to the third pillar of a stable banking sector: market discipline. With all this talk about rules and supervision, we must not forget that banks operate in a market economy. So why not use market forces to keep risks in check?
There is nothing like the prospect of failure and financial loss to keep a lid on risk-taking. And this, as a general rule, is the essence of market discipline. Investors who stand to lose their own money will be more cautious than investors who can offload losses onto someone else.
But, as I said earlier, markets have not always exerted discipline. During the crisis, bank failures were rare. But why?
Well, first, there was a lack of tools to resolve failing banks – particularly across borders. And second, there were fears of contagion and a meltdown of the entire financial system.
So, in the end, governments often stepped in and propped up failing banks with public funds. There was no chance for the market to impose discipline.
This has changed. In Europe, we now have a single resolution mechanism for banks. It was designed to ensure that banks can fail in an orderly manner without damaging the financial system. This new mechanism has recently passed its first series of tests. Generally speaking, it was successful, although some adjustments are still needed.
In any case, we have taken a big step towards more market discipline in the European banking sector.
To sum up, a stable banking sector is supported by three pillars: regulation, supervision and market discipline. Since the crisis, we have made these three pillars stronger. Now we have to make sure that they remain in good shape and share the load in a balanced way. Otherwise, we will not reap all the benefits of the transformation that has taken place in banking supervision.
Thank you for your attention.
European Central Bank
Directorate General Communications
- Sonnemannstrasse 20
- 60314 Frankfurt am Main, Germany
- +49 69 1344 7455
Reproduction is permitted provided that the source is acknowledged.Media contacts