ECB supervision at five: re-charting the route
Speech by Ignazio Angeloni, Member of the Supervisory Board of the ECB, at the Financial Times – Fitch Global Banking Conference, London, 2 May 2018
It is a pleasure to be here today, and I would like to thank the organisers for inviting me.
One of the things for which the British are admired worldwide is their sailing tradition. Another is their sense of humour, which led one of them to say that sailing is, in essence, like standing in a cold shower, with your clothes on, while ripping up banknotes. Those of you familiar with this expensive sport will probably know what “rounding the windward mark” means: after the first leg of a race, in which the boats sail windward and spread out across the water, they all regroup to round the first mark. This is the pivotal moment that reveals how the boats are performing and sets the tone for the rest of the competition. It is also the point at which the routes of the boats have to be re-charted.
I mention this not only because I myself share – or rather, I should say, shared in the past – a passion for active sailing, but because I feel that this idea now applies metaphorically to European banking supervision as well. As you may remember, the Single Supervisory Mechanism (or the SSM, as we call it) started to operate at the beginning of 2014. This means that at the end of this year it will pass the five-year mark. This is, in many ways, a natural time at which one may assess performance and consider future prospects and strategies with the benefit of a little hindsight. In my remarks today, I will look back at what ECB Banking Supervision has accomplished, as well as what may still be missing or incomplete. I will then venture into the future and try to identify some of the challenges and developments that lie ahead.
You may wonder why I have chosen to talk about this subject, one that may appear rather inward-looking and continental, at an event here in London, a city which is not only outside our banking union, but will also soon be outside the European Union. The reason is that I believe that Brexit will require more cooperation and exchange between authorities and market participants across the Channel, not less, although the modalities will be different. Brexit will not mark the end of cross-border banking and financial activities between the United Kingdom and the Continent, but it will make them more complex, more costly and, at least initially, more uncertain. We in the SSM get a sense of that uncertainty in our daily interactions with banks that are relocating operations from the City to the euro area, banks that will soon be supervised by us. It is essential that we improve their perception of how ECB Banking Supervision works, as well as, of course, our familiarity with how they operate. The rift that Brexit creates must be spanned by a bridge of understanding. There is much to gain on both sides.
Strategic directions taken by ECB Banking Supervision
I will now briefly, and selectively, sum up the main directions taken by the ECB’s banking supervision arm since its inception. I will organise my comments under three headings:
- Achieving a sounder capital base
- Reducing credit risks
- Improving the quality of supervision – a catch-all that includes several actions
This is an extreme simplification. I will say nothing, for example, about the initial set-up. Five years ago, ECB supervision consisted of a few dozen individuals scattered around half-empty offices. Today, it counts over 1,000 experienced supervisors, hosted in a dedicated and fully equipped high-rise in central Frankfurt, organised in well-structured work lines, fully integrated with the national supervisors. This organisational achievement, unprecedented in speed and outcome, has won the praise of both the IMF and European auditors recently, but I will not go into that today. Nor will I discuss what we do regarding supervision of the smaller banks, whose daily supervision remains a national responsibility and for which the ECB performs only an indirect oversight function.
Achieving a sounder capital base
At the start, the overriding priority of the newly established SSM was to re-build confidence in European banks. The widespread view, especially across the Atlantic, was that European banks were not “credible”; capital was deemed insufficient, perhaps overestimated after the crisis as a result of hidden losses on balance sheets. Confidence needed to be re-established as a matter of urgency.
This started with a “comprehensive assessment”, a first check prior to assuming a formal supervisory role. We knew, at the time, that this initial step would not suffice. Indeed, in terms of capital strengthening, much more was achieved in the following years. A few numbers can illustrate this. Between end-2014 and end-2017, the CET1 capital ratio for significant banks directly supervised by the ECB rose on average by over 3 percentage points, to 14.6%, and the total capital ratio increased by over 4 percentage points to 18.1%. By focusing its requirements mainly on CET1, the capital with the best loss-absorbing capacity, the ECB’s supervision brought about an improvement in the quality of capital as well.
It is important to note that this progress was made first and foremost through actual capital increases, not through deleveraging or adjustments in the risk weights. I discussed this in greater detail, also providing figures, in a recent presentation I gave in Washington, which is available on the ECB’s website.
Another landmark concerned the method used to determine our capital demands. For this purpose we set up a new risk assessment methodology, the Supervisory Review and Evaluation Process, or SREP. Every year, the SREP aggregates all known risk factors, bringing in quantitative and qualitative information which is then distilled into scores measuring the overall riskiness of banks. This annual exercise determines the capital and liquidity requirements that the bank will need to meet the following year. The SREP makes the process objective and systematic, promotes a level playing field and facilitates the understanding and acceptance of supervisory decisions.
I should mention also that the strengthening of balance sheets was not limited to capital. Between end-2014 and end-2017, the liquidity coverage ratio of the banks we supervise rose on average by 16.4 percentage points to 143.6%, and the net stable funding ratio rose by 11 percentage points to 113.4%.
A natural question to ask now is: is this enough? Have we reached a degree of capitalisation that is appropriate or acceptable for our banks? This is not an easy question to answer. Banking research does not provide conclusive answers on what the “optimal” bank capital is. That said, it is noteworthy that, while the amount of new capital demanded by the European supervisor on average has not increased since 2016, the actual capital ratio of the system has naturally continued to increase, an indication that the market still requires banks to enhance their solvency standards. International comparisons are also informative; CET1 ratios of euro area banks have continued to increase on average over the period 2014-17, while that of US banks, whose level was higher to begin with, remained broadly unchanged.
Reducing credit risks in the system
Another priority for ECB Banking Supervision has been to tackle the credit risks, posed by the high level of non-performing loans (NPLs). As in the early days of the euro the ECB was at times caricatured as being “obsessed with inflation”, the banking supervision arm of the ECB has often been portrayed as being “obsessed with non-performing loans”, supposedly neglecting the fact that more bank lending is needed to support growth.
Non-performing loans placed a huge burden on euro area banks, amounting to a stock in excess of €950 billion when the ECB formally assumed its supervisory duties at the end of 2014, or almost 10% of euro area GDP at the time.
We responded to this challenge in two ways. First, in early 2017, the ECB published a qualitative guidance to banks on how they should tackle non-performing loans, addressing key aspects of strategy, governance and operations. A key recommendation was that banks should set up internal structures to correctly measure the true scale of their NPL problem, with reliable data, and subsequently to manage the disposal process, by creating independent units tasked with this purpose and accountable to top management. The guidance also requested that banks define and implement plans to reduce NPLs in a realistic but ambitious manner. Second, in early 2018 the ECB published an addendum to the guidance, indicating supervisory expectations with regard to prudential provisions for new NPLs. The aim of the addendum is to avoid new NPLs piling up in future by fostering timely provisioning practices. The supervisory expectations provide a starting point for the supervisory dialogue between the ECB and each bank. The need for further provisioning expectations on legacy NPLs is currently being assessed.
These measures have helped to put NPLs on a steep downward trend. The gross NPL ratio of the banks we supervise stood at 4.9% at the end of 2017 on a weighted average basis, against 7.6% at the end of 2014; this amounts to a reduction in the total outstanding NPL stock of approximately 25%. Net of provisions, the NPL ratio declined by 1.8 percentage points to 2.7% over the same period. The progress has been even greater in the larger euro area economies with higher NPL levels. For example, the gross NPL ratio for the significant banks we supervise in Spain between 2014 and 2017 dropped from 8% to 4.5%. In Italy, the gross NPL ratio of the significant banks we supervise fell by 6 percentage points, to 11.1% at end-2017.
The ECB’s action plan on NPLs has been supported by constant coordination with the European Banking Authority and the European Commission, but has faced opposition from some parts of the banking sector. Some claimed that the ECB had overstepped its mandate and entered the domain of banking regulation. In fact, as was repeatedly clarified, our measures on NPLs amount to “Pillar 2” supervisory measures, set for individual banks. The European Commission has complemented on the legislative side, proposing legislation to address NPLs with mandatory requirements (i.e. “Pillar 1”).
Improving the overall quality of supervision
A third priority has been to improve the overall quality of supervision. Measures have been undertaken in various areas, in particular fostering harmonisation, convergence, transparency and even-handedness across the banking system. Let me mention a few initiatives.
First, there has been a push to harmonise the discretion left to the national supervisors by the European legislator. Through the ECB guidelines on “options and discretions” released in November 2016 agreement was reached among all 19 national supervisors participating in the SSM on how to apply the flexibility granted by the legislation. This was an important step to level the playing field across different countries.
Second, the ECB published a guide on criteria followed to assess the suitability of banks’ managers and administrators. Released in May 2017, this guide aims to increase transparency by outlining the criteria to which bank managers and administrators are subject as part of the normal vetting process. Fit-and-proper assessment is an important supervisory chapter; our experience shows that bank weaknesses can often be traced back to malfunctioning governing bodies.
Finally, in May 2017 the ECB released guidance on leveraged transactions. This provides a definition of such transactions and also outlines expectations regarding their proper risk management and reporting requirements.
Where now for ECB Banking Supervision?
Let me now leave the reassuring realm of retrospection and move to more treacherous terrain, that of speculation about the future. I need to reiterate at this point that the views expressed here are my own and should not automatically be attributed to anyone else.
I will briefly elaborate on four strategic directions, partly interrelated, that I regard as important for ECB Banking Supervision in the foreseeable future.
First: further enhancing the quality of supervision
Professional expertise and mission attachment were achieved rather quickly by ECB supervision; first of all, as I mentioned, the ECB was able to attract high-quality staff and inherit the expertise of able and experienced national supervisors; secondly the clarity of its statutory mandate left little doubt about where the ECB would be heading when exercising its responsibility as supervisor.
Fairness, transparency and independence are more complex requirements. They are particularly necessary for an international organisation, because its authority spans different business and supervisory cultures, but they are also more difficult to interpret and apply.
Supervisory fairness must ensure equal rules while taking into account differences in local conditions and business models. The ECB has done everything in its power to harmonise rules and their application, through the guidance on options and discretions, but the task remains incomplete. Progress is hampered by uneven legislation, which leaves ample room for further national discretions and rulings. Unlevel playing field and lower quality of supervision are more likely to arise in “softer” areas that have an indirect critical influence on bank performance, such as governance, internal controls, fitness and propriety of managers and administrators, where national law still rules, or in areas where the ECB has limited control and resources, such as on-site inspection activities. Progress in this complex and multifaceted area requires both action by the ECB (also through its moral suasion) and the support of European legislators.
Transparency is an area in which major advances have been made partly as a result of the clear prescriptions of the ECB supervision charter. However, this is still a moving frontier, where further progress is possible, allowing for more disclosure in supervisory methodologies and certain elements of supervisory judgement. The style and intensity of communication has changed radically in recent years in all areas of public policy. A similar evolution can be expected in banking supervision. Certain specific constraints need to respected, especially when dealing with proprietary information of listed companies, but I do not believe these obstacles are prohibitive.
Finally, the ECB’s supervision enjoys a high degree of operational independence, facilitated by our distance from national public and private interests. Supervisory capture is effectively being kept under control, also by a suitable rotation within staff teams. Nonetheless, this is an area requiring constant attention.
Second: strengthening the focus on market risks
As I explained, credit risks were a major focus of attention in the early years. Exiting from a long and deep economic recession, European banks were weakened by large volumes of NPLs, exacerbated, in a number of cases, by weak lending standards, lack of awareness of the problem and insufficient ability to deal with it. Certain bad habits needed to be corrected by firm supervisory guidance. But the landscape is now evolving. Major results have been achieved, and the progress continues. The main strategic decisions have been taken. The focus moves to implementation.
While this is happening, two factors suggest that some supervisory attention should be redirected onto market risks.
First, the expected normalisation of interest rates is already accompanied by higher volatility in the financial markets. More may come, and banks are not fully prepared for it. The long period of abundant liquidity and sub-zero rates has incentivised risk-taking and carry trades that are only sustainable in a low rate environment. Recent ECB analysis of interest rate risk in the banking book has revealed that internal models tend to be based solely on a period of decreasing interest rates. Some banks are still burdened by pre-crisis exposures, unprofitable and illiquid.
Second, Brexit will be a game changer as regards the mix of business models in the euro area. Global players relocating to the Continent have a higher share of market exposures, on average, than euro area banks. Increased competition and the prospective launch of a capital markets union in the EU will force euro area banks to rethink their own traditional models as well. While Brexit undoubtedly represents an opportunity to develop the financial structures in the continent, it also entails risks. The supervisor needs to take this into account.
We have already taken some steps in this area. For example, we are looking into banks’ positions regarding instruments which are less liquid and harder to value: the “Level 2” and “Level 3” exposures. In some cases this requires targeted analysis of fair-value financial instruments at selected banks, including long and complex on-site inspections to ensure that such instruments are correctly classified from an accounting point of view and that banks’ internal risk controls function appropriately. The ECB is conducting a review of internal risk models (the targeted review of internal models, or TRIM). This entails a systematic assessment of all the risk models used by the banks we supervise, including models for market risk. TRIM was launched in 2016 and is a multi-year undertaking scheduled to finish in 2020.
In the 2016 EU-wide stress test, market risks were a component of the overall short-term capital impact under the adverse scenario. Last year, the ECB conducted an analysis of how sensitive balance sheets are to interest rate risk in the banking book. These analyses need to be pursued further and broadened. Market risks in banking and trading books should be captured more systematically in stress tests and other supervisory analysis. Attention should also be paid to banks’ robustness when faced with dysfunctional market conditions, including illiquidity and counterparty risks. Risk cultures, governance and controls, and checks and balances within banks should all become more central to routine and targeted supervisory reviews.
Third: strengthening macroprudential linkages
ECB Banking Supervision should also find ways to improve its linkage with macroprudential policy. The legal framework sets certain boundaries and modalities for this to happen. The SSM Regulation of 2013 established the ECB as a macroprudential authority, but only for a limited set of instruments covered by EU law (risk weights on certain exposures, liquidity requirements, own funds and capital buffers). Borrower-based measures, such as loan-to-value and loan-to-deposit ratio caps, often recognised as having a greater impact, remain with national authorities. For the instruments set out in EU law, the ECB can only “top up” (i.e. tighten) the policy stance – a decision that the ECB has so far never taken. Coordination among stakeholders involves complex and cumbersome notification and consultation procedures. Under the EU macroprudential framework, the ECB receives a constant stream of notifications of national decisions, mainly referring to countercyclical and systemic risk buffers, many of which refer to decisions not to use the instrument in question.
Has the complexity of the framework and its predominant national orientation been an obstacle to timely decision-making or proper EU-wide focus? The European Commission has recently reviewed the framework and sent to the other European legislators (Council and Parliament) a proposal broadly confirming the existing approach. This does not prevent the ECB from engaging further in a number of areas. Let me mention some of them now.
First, commonly agreed methodologies need to be used to systematically assess the consistency of capital buffers applied to “other systemically important institutions”. While national authorities set these buffers, a common analytical cross-check would give more structure and discipline to the process.
Second, a similar approach should be followed for countercyclical buffers, which have been used very rarely in the present economic and credit upswing.
Third, the ECB needs to develop more systematic analyses of the broader impact of its microprudential decisions. The ECB has well-developed research and financial stability functions, alongside a microprudential supervision arm. Synergies between the two approaches should be exploited fully, and the analytical functions of ECB Banking Supervision could be further developed.
Fourth: further developing crisis management and prevention frameworks
The crisis management framework of the banking union has, all in all, performed adequately: the few cases of bank failures were addressed smoothly, without disruptions or contagion – though not without occasional pain for some stakeholders. Concerns occasionally expressed about the complexity of the framework and the many actors involved (central banks and supervisors, both national and European, plus resolution and competition authorities) have not been borne out.
Nonetheless, two areas need to be flagged for further improvement. One pertains to regulation, the other to supervision.
On the regulatory side, state aid and resolution arrangements issued at different times may lead to contradictory outcomes. For example, national winding-up proceedings, which are not harmonised at European level, may be more favourable for some creditors (senior bondholders, non-covered depositors) than European resolution. This loophole creates an incentive to bypass European procedures. Legislation also needs strengthening in the area governing “early intervention” actions undertaken by the supervisor in the initial phases of a crisis. Early intervention should mark a clear break from normal supervision, entailing more intrusive actions and more active involvement of the resolution authority. This is not the case in the existing legislation. Finally, bank moratoria are not envisaged in European legislation and can only be imposed in some national jurisdictions, with modalities and limits that differ from country to country.
On the supervisory side, our experience highlights the importance of conduct risks as potential triggers of bank crises. Money laundering has recently come to the fore. National anti-money laundering frameworks are not harmonised across the EU, and their remits are not aligned with prudential supervision. A European authority for combating money laundering, distinct from but actively cooperating with the ECB supervision, remains the desirable goal in the medium term. Under present arrangements, close cooperation between the ECB and the national authorities involved is essential. We are studying ways to enhance such cooperation, essentially by intensifying information exchanges on relevant findings that emerge in the fulfilment of the respective responsibilities.
The first five years of ECB Banking Supervision are coming to an end; that offers us an opportunity to review our performance and, potentially, to plan for new directions to explore in future.
The advent of ECB Banking Supervision marked a quantum leap forward in the quality of banking supervision on the Continent. The key strategic directions adopted since its inception were fundamentally correct and very successful. I admit that I am not entirely free from a conflict of interest in expressing this judgement; fortunately other, more neutral parties share my view.
I have also argued that some elements of the prospective future environment, including Brexit, suggest some new directions may be taken by ECB supervision. I hope that the views I have shared with you today will contribute to a broader reflection on how best to develop the European banking union, also in the interests of our UK partners.
Thank you for your attention.
 I am grateful to Francisco Ramon-Ballester for preparing a first draft of this speech. I am solely responsible for the views expressed here and for any errors.
 The SSM Regulation, which conferred supervisory tasks on the ECB, entered into force on 3 November 2013. The results of the ECB’s “comprehensive assessment”, which was a first check on banks prior to the formal assumption of supervisory duties, were released on 26 October 2014. On 4 November 2014, the ECB assumed supervisory responsibility for banks in the countries participating in the SSM.
 The role of the assessment in re-establishing confidence was stressed in public communication. See, for example, the ECB SSM Press Briefing on the comprehensive assessment on 23 October 2013, available at https://www.ecb.europa.eu.
 See “How real is Europe’s banking union?”, presentation by Ignazio Angeloni, Member of the Supervisory Board of the ECB, at the Peterson Institute for International Economics, Washington D.C., 19 April 2018. Available at https://www.bankingsupervision.europa.eu.
 It is interesting to note, in this respect, that the correlation between capital requirements imposed on Pillar 2 and banks’ overall SREP scores rising to 82% in 2017, from just 40% in 2014. This essentially means that the internal consistency of the process has increased.
 I discussed some research-related issues in “Banking supervision and the SSM: five questions on which research can help”, speech at the Centre for Economic Policy Research’s Financial Regulation Initiative Conference, organised by Imperial College Business School/CEPR, London, 30 September 2015. Available at https://www.bankingsupervision.europa.eu.
 This was partly the result of the introduction, in 2017, of a new notion of “pillar 2 guidance” (P2G), placed on top of the more traditional “pillar 2 requirement”. The P2G is a medium-run target, not binding in the short term.
 Based on a sample of 41 US banks and 33 euro area banks for which data is consistently available. Data source: SNL Financial.
 Article 1 of the SSM Regulation confers on the ECB “specific tasks concerning policies relating to the prudential supervision of credit institutions, with a view to contributing to the safety and soundness of credit institutions and the stability of the financial system within the Union and each Member State, with full regard and duty of care for the unity and integrity of the internal market based on equal treatment of credit institutions with a view to preventing regulatory arbitrage.” See Council Regulation (EU) No 1024/2013 of 15 October 2013 conferring specific tasks on the European Central Bank concerning policies relating to the prudential supervision of credit institutions, available at https://eur-lex.europa.eu.
 The recitals to the SSM Regulation contain a number of provisos in this regard, for example noting that “the ECB should comply with the principles of due process and transparency”. See Council Regulation (EU) No 1024/2013 of 15 October 2013 conferring specific tasks on the European Central Bank concerning policies relating to the prudential supervision of credit institutions, available at https://eur-lex.europa.eu.
 This is also mentioned in the ECB’s supervisory priorities for 2018, available at https://www.bankingsupervision.europa.eu/banking/priorities/html/index.en.html
 In this context, it should be noted that the European Commission’s communication on state aid rules to support banks (known as the “Banking Communication”) dates back to 2013, whereas the Bank Resolution and Recovery Directive (BRRD) entered into force in 2015 and 2016, such that the two pieces of legislation are not fully harmonised.