European banks and the banking union
Remarks by Ignazio Angeloni, Member of the ECB’s Supervisory Board, at the European American Economic Forum organised by Euronext and the European American Chamber of Commerce,
New York, 7 June 2016
Ladies and gentlemen,
It is a pleasure to be here in New York today and to participate in this panel. I hope that my presence, together with that of the other European speakers, will contribute to a better understanding of the reforms being undertaken in the banking sector on the other side of the Atlantic.
In view of the limited time, I will focus my remarks on three issues: first, recent developments in the euro area banking sector; second, the activities of the recently established ECB-based European banking supervision; and third, the steps that remain in order to complete the European banking union.
Recent bank performance in the euro area
The recent stock market developments at global level and in the euro area in particular have not been favourable to banks. In last 12 months, euro area bank stocks lost roughly 30% of their value, underperforming by far relative to both a general euro area stock index and a global index of developed countries’ banks. Most of the decline was concentrated at the beginning of this year. European bank stocks are currently trading some 40% below their book value. However, in order to better understand them, these developments must be looked at in a broader context. In the three years from mid-2012 to mid-2015 the Euro Stoxx Banks index rose by two-thirds, outperforming both of the other indices I just mentioned.
Interestingly, those three years coincided with the launch of the European banking union and the first steps in its implementation. In June 2012, political leaders announced the creation of a European banking supervision. The new Single Supervisory Mechanism (SSM) was to be hosted by the European Central Bank (ECB) and would be responsible for supervising all the main banks in the 19 member countries of the euro area. The ensuing preparation phase covered roughly the following two years, and in November 2014 the SSM started operations. Nevertheless, there were certainly other factors that influenced bank stocks at that time.
The headwinds that euro area banks currently face, as well as their future prospects and the broader implications for the economy, can be better understood by considering developments in three areas: credit market developments, bank profitability and non-performing loans.
Recent data on credit market developments indicate that bank lending in the euro area is recovering, albeit from low levels. In April, loans to households grew at 1.5% year-on-year, and loans to non-financial corporation grew by 1.2%. This subdued credit growth reflects both supply and demand factors. At the same time, the ECB’s bank lending survey indicates that banks across the euro area are easing their lending policies. To me, this holds an important message: the strengthening of the capital position of banks has not resulted in a “credit crunch”, as some had feared. The expansionary stance of the ECB’s monetary policy and the strategy pursued by European banking supervision have together facilitated, for the euro area banking sector as a whole, the joint achievement of sounder balance sheets and easier credit conditions.
Bank profitability has recovered recently: the return on equity (ROE, ratio of net after-tax returns to common equity) of the banks directly supervised by the ECB increased from 2.8% in 2014 to 4.5% in 2015. These numbers are encouraging, after several years of low or even negative returns, but compare less favourably with the pre-crisis levels. Just before 2007, the ROE of the biggest euro area banks was in the order of 15%. Lower bank profitability reflects multiple factors, some of which are proving rather persistent. The first factor is the low level of interest rates, which compresses intermediation spreads, a major source of income, particularly for traditional retail banks. The second factor is the high level of loan-loss provisions – a point to which I shall return. Some of the largest banks are also burdened by substantial litigation costs. And finally, many euro area banks, both large and small, are traditionally characterised by a burdensome and rigid cost structure. While these factors need considering, one should also recognise that a lower return on equity now can to some extent be expected, given that the new post-crisis environment is characterised by very low or even negative nominal returns throughout the financial market.
Another relevant issue for euro area banks is the large stock of non-performing loans (NPLs). They clog the asset side of banks’ balance sheets and weigh on their profitability, and thus hamper the banks’ ability to provide new lending to the economy. Banks bearing a heavy NPL burden become risk-averse and less willing to lend. In 2015, banks in the euro area made progress in tackling non-performing loans by improving their processes and increasing provisioning levels. The NPL ratio, based on harmonised definitions, decreased by 0.7% in 2015 to 7% of total assets, while the coverage ratio (provisions as a ratio to NPLs) increased by 1.3%, to 45.3%. These numbers are averages, however, and performance varies widely across countries and individual banks. In particular, one in four large euro area banks had an NPL ratio above 13% as of end-2015; one in ten had an NPL ratio above 26%.
The activities of the Single Supervisory Mechanism
Let me now turn to the second part of my remarks: the activities of European banking supervision, the Single Supervisory Mechanism (SSM).
Since November 2014, the ECB has been directly supervising the largest and systemically more relevant banks in the euro area. The remaining small and medium-sized banks remain under the direct supervision of national authorities, under the oversight of the ECB.
Prior to assuming supervisory responsibilities, the ECB conducted an in-depth review of the balance sheets of those banks it was about to start supervising. This “comprehensive assessment”, as we called it, consisted of two parts: an asset quality review, which analysed the main components of the asset side of the banks’ balance sheets, and a stress test, which analysed the resilience of banks to economic shocks. The comprehensive assessment provided an invaluable amount of information which helped the ECB to understand the risks and challenges of the euro area banking sector. The assessment, combined with the competitive pressures in banking markets, also became a powerful catalyst for banks to strengthen their balance sheets far beyond the strict supervisory requirements. The average Common Equity Tier 1 ratio for those banks directly supervised by the ECB has increased from approximately 9% in early 2012 to over 13% now.
However, the comprehensive assessment deliberately focused only on a few drivers of bank risk, namely the quality of assets and the sensitivity to stress. Therefore, it was not intended to produce a final and definitive appraisal of the banks. This explains why the ECB subsequently imposed some (limited) additional requirements, even on banks that had successfully passed the assessment. Moreover, as international and European regulations have evolved, with new prudential instruments being phased in, the liquidity and funding structures of banks have more recently come under closer supervisory scrutiny as well.
Currently, a central concern of the ECB in its supervision function is to ensure that the business models of banks are sustainable, taking into account the prevailing and foreseeable economic conditions. At present, there are two main challenges to the business models, both of which I already touched upon. The first challenge is the low level of interest rates. Low interest rates have persisted for some time now – not only in Europe. Since it is unlikely that interest rates will rise again soon, banks need to adapt to the new environment by containing costs and developing alternative sources of income. The increase in profitability in 2015 was encouraging but it remains to be seen whether this is a sustainable development. The second challenge is the large stock of non-performing loans.
Like other supervisors in Europe and elsewhere, the ECB assesses bank risks and sets prudential requirements for individual banks on an annual cycle basis. It does so through the so-called Supervisory Review and Evaluation Process, or SREP. In 2015, the SREP was, for the first time, conducted according to a unified methodology throughout the euro area. The methodology combines quantitative and qualitative elements and treats all banks consistently, while accounting for different business models. In a nutshell, the SREP consists of four components: an assessment of the bank’s business model, an analysis of its internal governance and risk management, an analysis of risks to the bank’s capital, and finally, an analysis of risks to the bank’s liquidity and funding. Each analysis relies on quantitative and qualitative tools. The quantitative assessment uses a broad range of data, including data on own funds, on large exposures and on credit and operational risk. The qualitative assessment involves judging things such as the quality of internal risk controls, risk culture and governance. That assessment combines expert knowledge with supervisory experience. Within the SSM there are dedicated supervisory teams (called Joint Supervisory Teams, or JSTs) responsible for providing this qualitative input.
As a result of the 2015 SREP, the supervisory capital requirements for those banks directly supervised by the ECB increased by 30 basis points on average compared with the previous year. In addition, the phasing-in of macroprudential buffers implied an additional increase of 20 basis points, making a total increase of 50 basis points. This moderate increase was deemed adequate and gives rise to a gradual transition towards the “fully loaded” Basel III requirements.
This year, we will further refine the SREP-methodology in order to make it more flexible, more responsive to risks and more effective. We are reflecting on the possibility of distinguishing two components insofar as the capital adequacy is concerned. One would be the Pillar 2 requirements proper, to be fulfilled immediately and maintained at all times. The second component would be more flexible, indicating to the bank the adequate level of capital to be maintained over a longer horizon. This second part would not constrain the use of internal resources by the bank for variable compensation or for remunerating equity of Additional Tier 1 (AT1) capital.
Since its inception, the SSM has also been active on the regulatory front. In principle, the SSM does not possess regulatory powers. In Europe, regulatory powers lie with the European Commission, together with the European Council and the European Parliament; for lower-level regulation and technical standards, they lie with the European Banking Authority. However, in practice, European banking law contains several elements of flexibility, which allow the supervisory authority considerable room for manoeuvre – so-called options and discretions.
The need to harmonise the exercise of these options and discretions and to level the regulatory playing field became apparent during the comprehensive assessment, when we observed several significant differences in the quantity and quality of capital across euro area banks which were not justified by underlying risks. Subsequently, the ECB identified the options and discretions that needed to be harmonised, and then developed a consistent SSM policy approach towards them. The process ended with the publication of the final legal text in March this year. The harmonisation of options and discretions for significant banks is a major milestone on the road to consistent supervisory practices. It means that banks will operate within the same framework and be treated consistently – both from a supervisory and regulatory point of view.
Completing the banking union
Since the financial crisis, the regulatory and supervisory framework in Europe has been overhauled, and the course has been set for a fully-fledged European banking union. The banking union road map envisages a gradual transfer to the euro area level of the functions performed by the national authorities in the areas of banking supervision, crisis management and deposit insurance.
The European banking union is built on three pillars.
The first pillar is the establishment of a single supervisory authority within the ECB. I have already referred to it extensively. The SSM, which comprises both the ECB and the national supervisory authorities of the participating countries, has a statutory responsibility to ensure the safety and soundness of banks, as well as their systemic stability. The SSM Regulation assigns to the ECB the necessary supervisory instruments to attain these goals. The ECB directly supervises the 130 largest and most systemic banking group, accounting for more than 80% of bank assets in the region and including close to 1,000 individual entities. The smaller institutions (well over 3,000 of them) continue to be supervised by the national supervisory authorities, subject to ECB oversight.
The second pillar of the banking union is the Single Resolution Mechanism (SRM), which aims to ensure the orderly resolution of failing banks. There is, of course, a close connection between supervision and resolution. No supervision can be effective, particularly while enforcing tougher requirements on weak banks, if there is no confidence that bank failures can be handled in an orderly fashion. In a multi-country setting such as ours, area-wide supervision is not compatible with national resolution powers, lest incentives be distorted. European supervision calls for European resolution. The two functions must carefully complement each other and carefully interact.
In January this year, the Single Resolution Board, located in Brussels, started operating. It is supported by a Single Resolution Fund, which is gradually being built up by banking sector contributions to ensure that adequate funding is available during the restructuring of a bank. Also this year, the European Bank Recovery and Resolution Directive entered fully into force, although there is still a small number of countries which have not yet, or only partially, transposed that Directive into national law.
This year is indeed very important for the SRM as two fundamental objectives will have to be reached. First, resolution plans for the most important banks under European supervision have to be prepared. Second, also for those banks, the minimum requirements for own funds and eligible liabilities, MREL for short, will have to be determined. The new legislation requires that for each bank, a minimum endowment of resources will need to be available at all times to ensure that banks have sufficient loss-absorbing capacity on their own balance sheets, if they need to be resolved. These requirements will be determined by the SRM in consultation with the ECB, taking into account the structure and characteristics of each bank as well as its resolution strategy. Further, MREL will be set in conformity with European law – the Bank Recovery and Resolution Directive – and the standard for Total Loss-Absorbing Capacity issued by the Financial Stability Board.
The Bank Recovery and Resolution Directive and the SRM aim to provide a coherent bank crisis management framework, which includes several elements: instruments and powers for supervisory and resolution authorities, burden-sharing arrangements, a single resolution authority in the banking union responsible for significant institutions and cross-border groups and, finally, a single resolution fund. Resolution authorities are required to adopt resolution plans and to remove ex ante any legal and operational impediments to resolvability. Since it typically involves the use of collective resources, resolution can only be enacted if a public interest exists, in the sense of the public interest being better protected than it would be under normal insolvency procedures. A condition in the resolution process is that no creditor should lose more than it would in the case of liquidation (the “no creditor worse off” principle).
The new framework implies a major shift of the potential burden of bank crises from taxpayers to the direct stakeholders of each bank – shareholders and creditors. Such a shift – politically driven by the perception that the recent crisis imposed major costs on ordinary citizens – will contribute, after proper adjustment of the financial system, to more balanced burden-sharing and to sounder incentives for investors and bank managers. The transition may be challenging, however, because the new regime requires all investors to be fully aware of the risk characteristics of the instruments they invest in. Adequate transparency and financial education are therefore essential. Both the Bank Resolution and Recovery Directive and the pre-existing state-aid framework provide for safeguards and flexibility to cater for the risks of financial instability and disproportionate effects.
This takes me to the third pillar of the banking union: deposit insurance. After the Great Depression of the 1930s, deposit insurance mechanisms were established in most advanced countries. The rationale for insuring bank deposits is quite evident: a system-wide deposit guarantee scheme ensures that each deposited euro is equivalent, regardless of the bank that received it, and reduces the risk of devastating bank runs, such as those which occurred repeatedly during crises in the past. Equivalence and safety are public goods which justify an intervention by the public sector, also in the form of a backstop. Again, and for similar reasons, area-wide supervision and resolution arrangements call for area-wide deposit insurance.
The Five Presidents’ Report entitled “Completing Europe’s Economic and Monetary Union”, published in June 2015, emphasised the need to complete the banking union with a European deposit insurance system. In November last year, the European Commission made a proposal for a European Deposit Insurance Scheme (EDIS). The idea is to build on the existing national schemes and then gradually move towards a European construction. Three steps are foreseen, with the starting point being a re-insurance mechanism built on top of the national schemes. The second step would see that re-insurance mechanism progressively evolving into a co-insurance arrangement. The third step would be taken in 2024. From then on, EDIS would fully insure all deposits in the euro area. As such, the Commission’s proposal combines clarity of purpose with gradual implementation.
While the rationale for a banking union comprising all three pillars is generally accepted, controversy has arisen recently regarding the timing of the third pillar, the European deposit insurance. Critics of the Commission’s proposal argue that a mutualised scheme cannot be put in place until the risks inherent in the national banking sectors have been contained and have become more homogeneous. In particular, the exposure of national banking sectors to their own national sovereigns (“home bias” in portfolio holdings) is seen as a fundamental source of asymmetry in the degree of risk facing national banking sectors. In this way, the debate on the completion of the banking union intersects with another thorny issue faced by global regulators, that of the prudential treatment of sovereign exposures.
While these concerns should be taken seriously, one must also recognise that an incomplete banking union, if allowed to persist for long, can easily become an additional source of uncertainty and risk. Policymakers should limit these risks by drawing up a coherent strategy, with clear goals and a reasonably fast implementation timeline. This, in essence, is the intent of the Commission’s proposal, and the ECB supports it. In its supervisory function, the ECB is fully engaged in reducing bank risks and making them more balanced, transparent and manageable.
Regarding the prudential treatment of sovereign exposures, there are indeed good reasons for considering a modification of the rules in the future. Under current banking law, banks’ exposures to their domestic sovereigns can be exempt both from risk weights and from large exposure limits. However, government bonds are far from being riskless – this has been amply demonstrated by the sovereign crisis in the euro area and by market valuations. Treating government bonds as risk-free is therefore not prudent, it sets the wrong incentives for banks’ investment decisions, and it unduly subsidises credit to governments at the expense of private borrowers and ultimately of economic growth.
Any regulatory change, however, should be mindful of the financial risks and should take into account the central role that sovereign debt instruments play in the financial sector and in the monetary policy process. Rigid limits should be avoided, whereas consideration could be given, for example, to more flexible approaches consisting of the application of gradually increasing risk weights to exposures concentrated on individual sovereigns, beyond certain thresholds. The result, ceteris paribus, need not be a decline in the total amount of sovereign bonds held by banks, but rather greater diversification by issuer. In any case, an appropriately long phasing-in period should be foreseen to avoid abrupt effects and be able to adjust in response to experience. The Basel Committee on Banking Supervision has started to reflect on the issue and this may lead to the establishment of an international standard. Any changes in Europe should, in our view, take place within that framework.