The Single Supervisory Mechanism after one year: the state of play and the challenges ahead
Speech by Danièle Nouy, Chair of the Supervisory Board of the Single Supervisory Mechanism,
Banca d’Italia conference “Micro and macroprudential banking supervision in the euro area”, at the Università Cattolica,
Milan, 24 November 2015
Ladies and Gentlemen,
Many thanks for the invitation to speak to you about our experience during the first year of the Single Supervisory Mechanism (SSM) and the challenges that lie ahead in the setting of this prestigious University. Let me first provide you with some numbers to give you an idea of the scope of our work. We are currently responsible for the direct supervision of 122 banking groups (possibly 130 next year). The balance sheets of these 122 banks account for €25 trillion euro in assets. We issued over 1,200 supervisory decisions in respect of those banks last year. However, what have we achieved so far and what are the main challenges ahead? This is what I will talk to you about today.
Where have we come from and what was achieved in the first year?
The roots of the SSM
Taking into account that many students are among us, it is worth reflecting for a moment on where we started. This is always important to keep in mind, especially for such a new set-up as the SSM. Here, we can choose between two different perspectives.
From the first perspective, one would see the financial crisis of 2007 as the main reason for developing a European banking union. The crisis showed that a monetary union without a banking union is fragile. This was further demonstrated by the subsequent sovereign debt crisis, with the emergence of the sovereign-bank nexus, which amplified the seriousness of the crisis in the euro area, destabilising our economies and ultimately our single currency. Europe had many supervisory authorities holding similar responsibilities, but operating in different national jurisdictions. This was problematic: supervisory authority stopped at national borders, but banks’ operations did not. Under this imperfect system, the costs of any failures were ultimately borne by taxpayers and vicious cycles developed. From this perspective, it is fair to say that the financial crisis was a catalyst for the creation of a European banking union in order to safeguard economic and financial stability in the euro area.
But, rather than a revolutionary decision, the creation of banking union was more of an evolution, albeit an “evolution at the speed of light”. And this is where the second perspective on the genesis of the SSM comes in. Even before the crisis, the need for more convergence in banking supervision was recognised. The Lamfalussy reforms initiated in the early 2000s permitted the development of a single rulebook. The Committee of European Banking Supervisors (CEBS) was established in 2004 to improve supervisory cooperation, including through the exchange of information between national supervisors. So long before the crisis, European prudential supervision was already on the path to further convergence.
Against this background, the harmonisation of supervision in Europe did not come out of the blue. The financial crisis functioned as a catalyst and was in itself the best proof of the need for further European convergence. Therefore, in 2011 the CEBS was replaced by the European Banking Authority (EBA), which was given greater regulatory powers with the ability to adopt binding technical standards and guidelines, and from 2012 the idea of a European banking union gained prominence within the euro area, ending up with the decision to launch the SSM. As you know, the banking union is based on a single supervisory, resolution and deposit insurance mechanism that complements the single monetary authority. The banking union is still a work in progress: the Single Resolution Board will become operational on 1 January 2016 and the deposit insurance mechanism is still being developed. I am glad to say, however, that the Single Supervisory Mechanism is fully functioning. This is in part precisely because we were able to build on foundations that had already been laid before the crisis.
The first year
Three weeks ago, we celebrated the first anniversary of the SSM. However, our work started well before 4 November 2014, as we had to set up an efficient mechanism for supervision, based on cooperation between the ECB and the 19 national supervisors. Staffing the organisation was the first challenge. We recruited highly skilled staff with diverse backgrounds, some of whom had a long track record at one of the national supervisory authorities and were able to bring experience about national practices to the SSM. We also recruited staff from within the ECB and brought in people from the private sector. We set up the joint supervisory teams, the JSTs, which are the core of our banking supervision. Each JST supervises one of the significant banks and is led by an ECB coordinator. Two thirds of the members of each JST come from the relevant national competent authority (NCA), while one third is made up of ECB staff. This results in the JST being based both in the country where the supervised bank is located and in Frankfurt. In addition to the JSTs, we created what we call horizontal functions. The work in these divisions focuses not on one bank, but on one theme, such as risk analysis, authorisations or sanctions. The horizontal divisions are essential for the development of common methodologies and for quality assurance. Lastly, we have supervisors who are dedicated to the indirect supervision of less significant institutions, or LSIs. The supervision of these banks is the direct responsibility of the national supervisors.
With the contours of the organisation becoming clearer and the staff recruitment process in full swing, we conducted a comprehensive assessment of the significant banks within the euro area. The results of this health check has provided a solid starting point for the SSM.
Together with the NCAs and the EBA we carried out both a stress test and an asset quality review. In successfully completing this exercise, we confirmed the high standards we had set ourselves. Equally important, the comprehensive assessment gave us a very valuable set of additional data on the state of play at the banks we were tasked with supervising. Based on these insights, we were able to draw up a strategy for the first year of supervision and decide what steps we would take to tackle the weaknesses we found in the balance sheets of these banks.
Of course, our first focus was on adequate capitalisation of all of the banks. This is a precondition for, and our contribution to, financial stability within the euro area. We also focused on ensuring that management had proper control of banks, so that risk management would be effective and banks would be able to provide the wider economy with financial services. By focusing on these themes we covered four of the five priorities we set out for our first year: business models and profitability, governance and risk appetite, capital adequacy and credit risk. Our fifth priority was cyber risk and data integrity, which I will come back to later.
Another major part of our work in this first year has been developing common methodologies. This did not come as a surprise to us. Almost everything we did in this first year, we did for the first time. And because from the start we wanted to ensure consistency, we had to recognise the need to develop methodologies that we could use again in the future. This was a challenge, as we had to harmonise 19 different national practices, which not only the national supervisors, but also the supervised entities had grown accustomed to. Some banks were not used to the scope of our data requests, for example. While understanding that these changes take time to get used to, we also firmly believed in the need to develop common methodologies. In full cooperation with the national supervisors we made significant progress in this area. We drafted our Supervisory Manual, an important tool that provides all of our supervisors, both in Frankfurt and around Europe, with a common set of procedures and practices which they can use to execute their work. Whether it is an assessment of the issuance of a Tier 2 instrument by a bank, a bank’s remuneration policy, or a fit and proper assessment, we now use the same procedures.
Thanks to this work on methodology, we were able to conduct the first round of the common supervisory review and evaluation process (SREP) in 2015. For the first time, all significant institutions in the euro area were assessed against a common benchmark. Quantitative and qualitative elements were combined through a constrained expert judgement approach, which ensures consistency, avoids forbearance and takes into account institutions’ specificities. Extensive peer comparisons and transversal analyses were possible on a wide scale for the first time, thus fully promoting the goals of the banking union.
More specifically, in preparing the Pillar 2 capital decisions, the supervisors first took into account persisting and developing risks faced by banks due to economic and market conditions in the euro area, such as credit and liquidity risk. Second, they pursued the objective of a smooth transition by banks from current capital positions towards a fully-loaded Basel III environment in 2019. Third, they did all the above while maintaining a level playing field, both within the SSM and with other major jurisdictions.
In 2015, we have also been playing a key role in managing the financial turmoil in Greece, which was triggered by a deadlock in the negotiations on the Greek programme last June. This had an immediate effect on Greek banks, which again suffered a very intense stress episode. We have monitored the situation closely and have always been prepared to take supervisory action where necessary. In the final political agreement reached shortly before summer we were given a role in contributing to the determination of the recapitalisation needs of Greek significant institutions. For this purpose, we conducted a comprehensive assessment of these banks, the results of which were published on 31 October.
I have given you a quick overview of our first year. We have staffed our JSTs, carried out the comprehensive assessment, built important parts of our common methodology and carried out our first common SREP. We already have a lot to be proud of, but at the same time, we have only just begun.
More harmonisation of regulation and supervision
There are still a lot of challenges ahead, most notably in terms of achieving the main goal of the SSM: ensuring fully harmonised prudential banking supervision in the euro area. As I just mentioned, with the single rule book we have taken an important step towards harmonising the prudential regulatory framework, but we are not quite there yet. The implementation of the Capital Requirements Directive (CRD) in national laws has led to many different interpretations of specific elements of the Directive. This poses a problem for the SSM supervision, because it is based on the transposition of directives into national laws. Gold plating and different national interpretations of CRD articles create an uneven playing field within the SSM, which goes directly against one of our main objectives.
In some cases, the CRD, the Capital Requirements Regulation (CRR) or delegated acts, grant participating Member States the explicit freedom to decide how to apply a specific requirement. I am talking here about the famous options and national discretions (ONDs), of which there are around 150. Some of these ONDs can be exercised by the national legislators, but in other cases this is delegated to the supervisory authorities. They cover a broad range of issues, such as intra-group exposure limits, or institutional protection schemes.
For these ONDs, we now have the possibility to agree on a common approach within the SSM. To this end, we established a high level group which identified around 120 ONDs that we could address during the first phase of our harmonisation project and made proposals on how to address them. We have set out our proposed treatment of these ONDs in a draft Regulation and Guide, which are currently going through a public consultation process. The implementation of the final Regulation will be a major improvement, but is at the same time only a first step. More ONDs have been identified, and we will consider a common approach to these in the near future.
As I said, it is not just the ONDs that create an uneven playing field, but in some cases also the way that primary legislation is interpreted and implemented in national law. Matters are further complicated by the fact that in some countries national pre-SSM supervisory practices were enshrined in national law, limiting the range of options for supervisors to carry out their work.
These are risky developments, because they create the very same fragmentation, national biases and opportunities for supervisory arbitrage that the SSM is supposed to do away with. For credible, unified and harmonised European prudential banking supervision, it cannot be the case, for example, that fit and proper requirements vary widely from one SSM jurisdiction to another. We are committed to achieving further supervisory harmonisation. It is, however, up to both the European and national legislators to improve regulatory harmonisation.
The banks: more resilience against adverse developments
Now I will turn to how the economic reality in which banks operate influences our supervision. Banks still face a number of challenges, including low profitability and the persistence of non-performing exposures on their balance sheets.
As regards the first, looking at recent developments, bank profitability slightly improved in the first half of 2015 and capital positions have further strengthened. But, despite this, the euro area banking system is continuing to struggle with low profitability and, for many banks, return on equity is hovering below the corresponding cost of equity. This is due to several factors. One is the relatively low economic growth levels faced by banks, which results in subdued loan growth. Within the euro area, growth levels are not only low, but uneven. There are significant differences in the state of the economies of euro area countries. Then there is the persistently low interest rate environment. In addition, banks are facing a high level of provisions. These developments will force a number of banks to fundamentally review their business models. The current economic environment might be an incentive for some banks to change their business models, in particular to rely less on traditional interest income generating business.
Non-performing exposures are also still a serious prudential challenge in some countries, including Italy. A number of factors underlie this problem, including general economic conditions, large stocks of legacy problem assets, particularly in countries most affected by the financial crisis, and sometimes flawed debt recovery frameworks. We as supervisors can address some of these factors, while others are outside our remit. In some countries, advances have been made towards a legal framework that is more conducive to effective non-performing loan resolution. The ECB has launched a dedicated workstream on this issue and is working with banks to develop and implement individual, tailored action plans.
Banks are also facing new regulatory developments. For example, banks have to remain on track with transitioning to a fully-loaded Basel III capital position in 2019. Next year, they will also have to implement new regulatory demands relating to “gone concern” situations. The total loss absorbing capacity (TLAC) requirement as developed by the Financial Stability Board is one example of a new regulatory development for the biggest banks in the euro area. In Europe, the Bank Recovery and Resolution Directive (BRRD) will come into force next January. The BRRD introduces a new minimum requirement for own funds and eligible liabilities (MREL).
Under these economic conditions, it is of vital importance that we ensure that banks have enough capital of good quality on their balance sheets. After the 2014 comprehensive assessment, we made sure that the capital gaps that were identified at some of the banks were filled. In our SREP assessments, we have taken into account the persisting and developing risks faced by banks owing to economic and market conditions in the euro area. We also factored in the transition path towards Basel III requirements.
The SREP decisions for 2016 are being finalised, but we can already say that the Pillar 2 capital requirements envisaged for the SSM’s significant institutions next year are slightly higher than in 2015 – by around 30 basis points on average. In addition to Pillar 2, the phasing-in of buffers requires around 20 additional basis points of capital. The JSTs began holding meetings with the banks in July to inform them about their SREP ratios. I believe that, once approved, the final results, together with the information on the most important aspects of the SSM methodology, will provide an indication of what the SSM’s Pillar 2 requirements are likely to be in steady state.
As you know, the results of our SREP have led to the criticism that the ECB is inconsistent. Some people argue that, while monetary policy is aimed at kick-starting the economy through quantitative easing, our banking supervision forces banks to hold more capital themselves via the SREP. The critics argue that setting capital requirements above the regulatory minimum smothers growth in the real economy; the same economic growth that the ECB is trying to foster through its monetary policy.
You will not be surprised to learn that I don’t agree. Let me tell you why. First, from a short-term perspective, it may feel intuitively correct to say that imposing additional capital requirements leads to less lending to the real economy. However, I doubt this. Most of our banks already have capital buffers that are above the minimum requirements, so their credit supply would not be significantly hit by our SREP decisions and the effect on the real economy will be limited. And even if a bank deleverages because of a SREP decision, it is reasonable to assume that it would do this primarily by shrinking its non-core activities. Its lending activities would not be affected initially. Moreover, a better capital ratio following a SREP decision would alleviate a bank’s funding costs, which should have a positive effect on lending to the real economy. Therefore, the short-term positive effects of not imposing capital requirements above the regulatory minimum are questionable.
However, the long-term negative implications of doing so are very clear. If banks don’t build up a cushion in the form of adequate capital buffers, they will be vulnerable to adverse development in the medium and long term. Uncertainty and low levels of trust in the banking sector hamper economic recovery. After all, the economy can’t rely on banks that are only able to perform well when the sun shines. This is why in our SREP decisions we require a bank to hold more capital than the regulatory minimum. Of course, it is not only a matter of the quantity of capital, but also of its quality. The work on ONDs is proving to be extremely useful for assessing the quality of capital that banks have on their balance sheets. In implementing the common SSM stances on the various ONDs, we will also make sure that the banks follow a reasonable time path towards reaching fully loaded capital ratios.
Having emphasised the importance of sufficient high-quality capital, please let me remind you that capital measures are just one of the tools we use to make sure our banks are resilient enough. From the crisis we have learned that banks will need to do more to ensure their resilience. The holistic view of the bank that we acquire through the SREP can be used by supervisors to counter the risks it faces. Imposing additional capital surcharges is an adequate mitigation measure for some of these risks, but not for all of them.
In our SREP we assess, for example, the bank’s governance, including its risk management and risk appetite. In this area, we found that important work still needs to be done. Consider, for example, the risks related to the outsourcing of services by banks to third parties, or the importance of adequate data aggregation and supply. It is vital that senior management and the board of a bank receive high-quality information on risk. The Basel Committee on Banking Supervision recently published principles for effective risk data aggregation and risk reporting which provide a good point of reference for banks to improve their practice. For us at the ECB, this will be a focal point for the coming year.
Another risk related to data integrity, but seemingly very different from other prudential risks, is cyber risk. This is a relatively new area of supervision that requires a different kind of expertise. However, it is by no means less important than any of the other prudential risks.
Previously, banks dealt with the risk that IT system failures could hamper their daily operations, trigger operational losses and cause damage to their reputations. But in today’s world, cyber risk also includes cyber attacks, the digital version of a classic bank robbery. These attacks can be the work of individuals, but also of criminal organisations. This year, we performed a cyber security review and identified the banks at which we wished to carry out on-site inspections. We are also setting up a process to closely monitor significant IT incidents at banks, which should allow us to get a good overview of trends and developments relating to cyber risk.
Governance, risk management and cyber risk are good examples of areas where additional capital requirements are not the first tool we use. Our goal of ensuring that banks have sufficient levels of high-quality capital will always be accompanied by a broader perspective on other prudential risks and measures to address these risks. You will recognise this broad perspective when reading our supervisory priorities for 2016, which we will shortly publish on our banking supervision website.
We have come a long way since the outbreak of the financial crisis in 2007. Today, supervisors are equipped with better tools to ensure that banks are resilient enough to withstand adverse developments. The harmonisation of prudential supervision has made a significant contribution to this. As I mentioned, we have achieved a number of important milestones in our first year: the Comprehensive Assessment and its follow-up; our first round of common methodology SREP decisions; and several thematic reviews. I have also pointed out that important work still remains to be done. On the harmonisation front, further regulatory and supervisory convergence is needed, because the single rulebook still allows too many differences within the SSM and the harmonisation of supervisory practices is not yet complete.
We all have a role to play here. National legislators and supervisors will have to be willing to change long-established practices. We at the ECB will constantly seek to promote further integration, in active cooperation with national supervisors. This is what we will focus on in the near future.
I thank you for your kind attention.