Caution should be the life of banking

Speech by Sabine Lautenschläger, Member of the Executive Board of the ECB and Vice-Chair of the Supervisory Board of the ECB, at the AFME Board Meeting in Frankfurt, 22 March 2017

Ladies and gentlemen,

Taking risks is at the heart of any healthy business, including banks. After all, it is the job of banks to take and allocate risks. But as Walter Bagehot said: “Adventure is the life of commerce, but caution is the life of banking”.

I agree with him.

Banks are far too important for the economy to indulge in adventures. They enable business by providing loans to companies or by helping them to tap the capital markets. And they also facilitate business by providing services in fields such as payments, mergers and acquisitions, and global trade.

Banks help the economy to grow and prosper. But they can also badly hurt it; we saw that in the financial crisis. That’s why banks need to be cautious. In practical terms that requires, among other things, sound risk management.

And this sets the scene for my short statement today. The topic is risk management. I will begin with some smaller, or rather bank-specific, risk management topics and end with a few big issues.

Sound risk management depends on good data. It also depends on firm-wide risk aggregation and reporting. Those who take decisions must be in the picture.

But are they? Well, in the crisis we learnt that decision-makers often were not. IT systems, data architecture and related processes were not up to the mark.

Against that backdrop, we assessed how banks aggregate data and report risks. The yardstick was a set of global best practices that were defined by the Basel Committee on Banking Supervision. These are known as the BCBS 239 principles.

The result of our review: many banks do not meet the global standards. As usual, we will therefore be very explicit on how these banks should improve. And we will closely monitor the progress made by the respective banks to ensure that they meet relevant standards.

Ladies and gentlemen, good risk data is one thing. But processing such data is another matter. And that leads me to my next topic: internal models. Many large banks use these models to calculate their risk-weighted assets, which then serve as a basis for calculating capital requirements. Over the past years, internal models have diverged greatly, not least due to the fact that supervisors applied different standards when authorising them.

In response, we have launched a project called TRIM. That stands for “targeted review of internal models”. It aims to ensure two things: first, that the playing field is level; second, that internal models produce reliable and comparable results. The calculation of risk-weighted assets should be driven by actual risks rather than by modelling choices. TRIM will help to raise trust in how banks calculate their risks.

Ladies and gentlemen, there are small risks and big risks. There are risks that affect a single bank and risks that affect the entire sector. And then there are risks that stretch across sectors and even countries. In that regard, there is one thing that comes to mind: Brexit.

Right now, it is still very unclear how the world will look after the UK has left the EU. Still, banks and supervisors must prepare for any scenario. The worst case would certainly be a “hard Brexit”. In that scenario, the UK would become a “third country” from the viewpoint of the EU.

As a result, UK banks might lose their European passport and, with it, access to the Single Market. This would affect many of the 40 UK banking groups that currently operate in the Single Market. They would then have to find another way in.

Most likely, they would seek to obtain a banking licence in an EU country to regain the passport. In the euro area, these licences are granted by the ECB. We obviously don’t care whether UK banks move to Frankfurt, Dublin, Paris or some other location in the euro area. What we care about are safe and sound banks.

We will therefore resist any supervisory or regulatory race to the bottom. Any bank that moves to the euro area will have to meet our standards – regardless of whether it comes from the UK or any other place.

Most importantly, we will not accept empty shell companies. All entities in the euro area must have adequate local risk management, sufficient local staff and operational independence. All that should be self-evident.

Banks which seek to permanently book all exposures back-to-back with another entity in London should change their plans. And so should banks that plan to book all exposures with a euro area entity while having their risk management somewhere else.

We are aware that it is a burden for UK banks to apply for a new licence in the EU. With a view to internal models, we would aim to be accommodating regarding the timing. There will be a transitional period in which new euro area entities might use internal models that have not yet been approved by the ECB.

But of course, there are conditions attached. First, the relevant models must have been approved by the UK’s Prudential Regulation Authority. We would also reflect on any comments on the quality of the models from our UK colleagues. Second, the banks must have applied for internal model approval in the euro area. The transitional period will cease as soon as we have approved or rejected the bank’s model application.

Ladies and gentlemen, sound banks are at the heart of a strong economy. Banks therefore should note Walter Bagehot’s words and remain cautious. It is our job as supervisors to help them do so and steer clear of risky adventures.

Thank you for your attention.