Common challenges for banks, regulators and supervisors
Speech by Ignazio Angeloni, Member of the Supervisory Board of the ECB, at the Adam Smith Society, Milan,15 May 2017
It is a pleasure to be here today, and I am grateful to the organisers for inviting me. 
The reference to Adam Smith in your Society’s name is inspiring, and also relevant to the topic of this meeting – namely, banking and its regulation. Among the classical economists, Smith is probably the most frequently quoted to this day. His enduring relevance and popularity are due to his having emphasised the social value of private initiative, an idea that is central in providing a theoretical underpinning to free-market economies. Less well known is the fact that Smith also advocated limits to individual economic freedom, for the latter to benefit society as a whole. In banking, Smith likened regulation to fire codes, which are needed even though they constrain individual freedom. Although he did not refer explicitly to “financial stability”, as we understand it today, he made it clear that regulating banking is necessary to safeguard the common good. Another of Smith’s ideas, rarely acknowledged but very pertinent today, is his insistence on the importance of ethics for the proper functioning of a market economy. Especially in finance, where informational problems are pervasive, the social value of private action depends not only on formal compliance with the rules but also on ethical behaviour by market participants. On another occasion I discussed how Smith’s ideas can help design ethical frameworks which are relevant to today’s discussions on banking reform.
Virtually unique among economic sectors, banking is subject, in addition to a regulatory framework – multiple levels of legislation, including technical and implementation standards – to an intermediate layer of control: banking supervision. In principle, the supervisor does not “make the rules”, but only ensures that the existing rules are properly applied. In doing so, however, it does exert an influence on the regulatory framework as well, first, because the legislator (which in Europe includes the European Commission, the EU Council and the European Parliament, complemented by the European Banking Authority for secondary legislation) naturally has frequent recourse to the advice of the supervisor, and second, because regulations normally allow for a degree of flexibility in their application. The ECB, the largest supervisor in Europe, is a relevant stakeholder in all regulatory discussion fora (not only European), and also has a margin of discretion in the way it applies regulations in its daily activities. However, it is also important to keep in mind that, just as regulations limit banks’ individual liberty in the way that Adam Smith had in mind, they limit the ECB’s supervisory power as well. I will return to this important point later.
I will briefly describe some of the challenges faced by supervision in what is known as “banking union” (which coincides, at present, with the euro area) and what the ECB is doing about them. They are, to a large extent, common challenges, implying that a high degree of understanding and cooperation between supervisors, regulators and the banks is needed. Considering that ECB supervision covers as many as 125 large banking groups and individual entities headquartered in or hosted by 19 different euro area countries, the scope of the challenges we face is very broad. To be concise, I will focus on three issues only: the repair of banks’ balance sheets (including our action regarding non-performing loans, or NPLs), our assessment and oversight of business models, and our efforts to promote an effective and balanced regulatory framework.
Repairing balance sheets
Even before the formal start of the Single Supervisory Mechanism (SSM) in November 2014, and ever since we conducted, during the course of 2014, our “comprehensive assessment” of the banks likely to fall under our supervision, a central aim of the SSM has been to reduce the risks built into banks’ balance sheets. That effort has achieved a fair degree of success, but is far from complete. The success is most easily visible in the improvement in solvency and other prudential parameters. The total capital ratio for the banks we supervise increased by 140 basis points in the period ranging from the second quarter of 2015 to end-2016. This increase is broadly spread across jurisdictions and bank categories. The quality of capital has increased as well, with the “high-loss absorption” component (CET1) now constituting an overwhelming share of total capital. Liquidity standards have also become more stringent, particularly with the entry into force of the new liquidity coverage regulation in October 2015. Here we have a good illustration of the three-pronged synergy I just mentioned: the progress was driven largely by supervisory action, but was supported decisively by a new and more effective body of regulation (the Capital Requirements Regulation, or CRR, and the Capital Requirements Directive, or CRD IV) and by the cooperation and initiative of the overwhelming majority of banks.
The ECB continues today to be engaged in further de-risking the banking system. Within the constellation of risks which banks face, credit-related ones constitute a major challenge, partly as a legacy from the crisis, partly deriving from inappropriate lending practices. High levels of NLPs hinder confidence in the banks and the recovery of the broader economy. Moreover, high NPL ratios tend to be concentrated in jurisdictions where economic activity and public finances are more fragile, implying a risk of negative feedbacks between banks and sovereigns. Helping banks dispose of large NPL portfolios, by setting ambitious but realistic targets is therefore in the interest of both the banks (microprudential purpose) and the economy in which they operate (macroprudential one). International experience has shown that unduly prolonging the presence of large NPL volumes can be a factor of economic stagnation and deflationary risk.
Recently, the ECB has issued a specific guidance to help banks tackle NPLs. After benefiting from a public consultation with the industry, the guidance addresses aspects related to NPL resolution strategy, governance and operations, soliciting banks to develop internal quantified action plans to reduce their non-performing exposures. As a result of the implementation of this programme, we expect a significant acceleration of NPL disposal. If the internal plans are credible, the confidence of investors and other stakeholders in the soundness of the banks involved will increase, even before the plans are fully implemented. The credibility is enhanced to the extent that the ECB’s supervisory action is actively internalised in the banks’ business decisions.
At the same time, we are not overlooking other sources of vulnerability. In particular, market risks are addressed at three levels. First, supervisory processes conducted by the Joint Supervisory Teams (JSTs) continuously review the relevance of these risks with regard to the balance sheets and the related governance and control processes within the banks. Second, specific attention is devoted to risks on assets and liabilities that are difficult to value, because of the lack of an active market. This has involved, on occasion, complex and lengthy on-site inspections and input by specialised experts. Third, an ongoing project aims at assessing the quality of internal risk models used by the banks, named “Targeted Review of Internal Models”, or TRIM. TRIM will bring more clarity on how banks using internal models to value their risk, notably (but not only) stemming from financial exposures. One obstacle is that, due to the large number of models employed by euro area banks and their complexity, the project takes time and uses significant resources of the ECB and the national authorities.
Internal models are used by banks, especially large ones to assess a variety of risks. Assessing their quality is particularly relevant when banks have to gauge market risks stemming from their so-called “level 3 positions”, assets and liabilities whose pricing cannot be based either on an active market or on indirectly observed prices. Determining whether the pricing model that produces the fair value is indeed reliable is thus a key question. Our initiative will complement and strengthen our regular supervisory engagement. It may also include, in the most relevant cases, targeted analyses of fair-value financial instruments at selected banks; on-site inspections to ensure the correct classification of fair-value instruments from an accounting point of view and the adequate functioning of banks’ internal risk control arrangements; and the inclusion of a wider assessment of risks to capital as part of our annual process of setting capital requirements for banks (the Supervisory Review and Examination Process, or SREP).
Another method to assess market risks is the stress test. This approach is less granular than the one obtained by analysing risk models, but it allows a holistic view of the impact of risky scenarios on the entire balance sheet and, potentially, on the banking system as a whole. In the last EU-wide stress test, conducted in 2016, market risks constituted a major component of the overall short-term capital impact from the adverse scenario. Losses on banks’ profit and loss account stemming from market risks (including counterparty credit risk and credit valuation adjustment) amounted in the same year to 53 billion euros for the aggregate of the banks analysed. Measured by the difference between adverse and baseline, in the first year the loss impact from the stress test was higher for market risks than for credit risks. However, credit risks tend to be more protracted, their effect lasting longer into the future.
The ECB has also been engaged, in recent months, in dealing with a few crisis situations regarding specific banks. I will not enter into details, as some of this work is ongoing. I would like to emphasise that when dealing with problem banks where State aid is involved, the ECB cooperates closely not only with the national authorities but also with the competition authority at the European Commission. The Single Resolution Board is also kept informed as well, as the law prescribes. When State aid is requested, as recently the case for some Italian banks, the restructuring plan falls under the responsibility of the Commission. The supervisor retains a role in providing specific input to the Commission, such as, for example, a solvency assessment and an estimate of the stress test impact, as prescribed by the Bank Recovery and Resolution Directive. Moreover, since the bank continues to be supervised by the ECB, the latter sets the prudential requirements needed for the entity to remain “safe and sound”. The responsibility for the ECB here is to ensure that the bank resulting from the operation is sufficiently strong from the outset and that its business plan is sustainable over time.
Promoting sustainable and adaptable business models
Another priority for ECB supervision concerns banks’ business models. Business model analysis constitutes one of the four macro-chapters which make up our SREP process (alongside governance, risks to capital and risks to liquidity). Importantly, the ECB does not prescribe specific business models to banks: this is a business decision. It rather aims to assess whether the business models proposed by the banks are sufficiently solid and resilient over time, even in adverse market conditions.
In recent years, the macroeconomic climate has been particularly adverse for banks, on account of the sluggish economic growth and the historically low level of nominal interest rates. Although the impact across individual banks has varied markedly, depending on their underlying asset and liability structure, and in spite of the fact that the ECB’s non-standard monetary policy measures have offset the associated downsides through different channels, there is no doubt that the traditional banking business – namely revenue generation through maturity transformation – has been hindered from both a supply (narrow margins) and demand (weak demand for credit) point of view. Moreover, the overcapacity of the sector, which has led to a steady concentration process in recent years, has put further pressure on banks’ earnings.
In spite of this environment, a significant number of banks have managed to consistently outperform their peers from the point of view of profitability. Remarkably, the most evident common denominator of the best performing banks, under this metric, is not the adoption of a specific business model, but their ability to contain costs. The large variance in average cost-to-income ratios across euro area banking jurisdictions also suggests that there is plenty of room for improvement in many cases. In order to shed further light on this matter, the ECB is carrying out a “thematic review” of bank profitability drivers, which also covers risks emanating from fintech and non-bank competition.
Looking forward, the normalisation of the interest rate structure should provide relief to their earnings. However, the transition and the ultimate effect is unlikely to be uniform across the bank population, depending on the composition of assets and liabilities, their maturity characteristics and technical forms. In order to shed light on this process, the stress-testing exercise of the ECB this year is focusing on interest rate risk in the banking book.
Towards an effective and level regulatory playing field
An effective and balanced legislative framework is essential for the proper functioning of the banking union. As I mentioned at the outset, the laws not only constrain the actions of banks, but also those of the supervisor. This is all the more true for the ECB, which exercises its supervisory function over a large number of countries, with partly different legal frameworks. The SSM Regulation (the charter issued by the European legislators in 2014 that established the goals, instruments and main rules of operation of the SSM) prescribes that in its supervisory decisions the ECB has to apply the European legislation and, when there is no European legislation directly applicable, the national legislation. The potential asymmetry stemming from this articulated legal basis is significant. It is also compounded by the fact that a large part of EU banking law is in the form of a Directive (the CRD IV), which enters into force only when transposed by a national legal act. Transposition laws are often an additional source of asymmetry, and may create a legal vacuum when transposition is delayed or is not complete. Another reason for complexity is that EU banking legislation predates the establishment of the banking union, and is therefore not equipped to deal with a higher degree of integration within the banking union, and with the presence of a centralised supervisory authority.
Within its realm of competence, ECB Banking Supervision has moved a step forward by harmonising the application of what are known as “options and discretions” in Union banking law. In both the CRR and the CRD IV, a number of provisions allow a degree of optionality, or leave some room for discretionary application, originally conceived to allow some adaptation to local market structures of preference by national supervisors. Consistently with the banking union, the ECB Supervisory Board has agreed on a harmonised approach to the exercise of such options and discretions, to achieve a more levelled playing field among its supervised banks.
This work is important but not sufficient. A good number of options and discretions are entrusted to Member States, and are hence outside the power of the banking supervisor. Additional asymmetries are generated by the transposition of the EU directives into national law. On both fronts, an opportunity for progress is offered by the ongoing review of European banking legislation, which should be concluded next year. Options and discretions on supervisory matters should be placed, to the largest extent possible, in the realm of banking supervisors. More extensive use should also be made of directly applicable provisions, in the form of regulations rather than directives.
An area where progress has been particularly difficult so far is that regarding the “fit and proper” assessments conducted by the ECB on the appointment of managers and key function holders. Here, the application of national laws has exposed numerous discrepancies across euro area jurisdictions. In response, the ECB has prepared a guide so that such assessments can be implemented in a harmonised manner. While this is a step forward, discrepancies are set to remain on account of both uneven transposition across countries and (in some cases) only a partial transposition of the CRD IV itself. In this area, we would like to see decisive convergence, so that the principles enshrined in the CRD IV, complemented by the new ECB policy, are fully effective. Often, malfunctioning of the governing bodies is at the origin of weaknesses and risks down the line. Attacking the problem at an early stage is essential.
Regarding the ongoing review of the CRR and CRD IV, the ECB has been engaged in close talks with the Commission with a view to ensuring that measures to strengthen and harmonise the ECB’s supervisory remit are undertaken. A formal ECB opinion on the Commission’s proposals will be published soon. This is the first review of European banking legislation since the start of the banking union. The opportunity to move forward towards more legal harmonisation and towards a fully effective supervisory function should not be missed.
Let me conclude this review on a positive note.
Recent data suggest that the euro area economy is improving, with the recovery becoming more broad-based across sectors and countries. Looking ahead, real growth in the euro area should firm further, at a pace slightly above previous expectations. According to the latest ECB staff macroeconomic projections, domestic demand should constitute the mainstay of growth. The latest projections by the European Commission confirm this trend.
This is good news for the banking sector. Banks’ balance sheets should benefit from a variety of channels, including the improvement in the condition of borrowers, a pick-up in credit demand, and a normalisation of term premia in financial markets. Interest rate risks, which will be the focus of our stress-testing exercise this year, may add some extra uncertainty, but are expected to remain contained.
While these more favourable prospects will help the reabsorption of NPLs, they also carry a risk of complacency; banks may be tempted to soften their commitment to balance sheet cleaning, in the hope that the improved economic conditions may “do the work for them” at a lower cost.
This would be a mistake. As I already mentioned, not all the NPLs we observe today originate from the recession; some can be explained by bank-specific factors. There are significant differences in banks’ credit performance and in the quality of their internal controls and governance practices, even between banks operating close to one another. There is therefore a need for continued effort, tailored to the specific situation of individual banks, to improve lending practices and to effectively dispose of existing NPLs. It should also be considered that while the recovery will not benefit all countries and all banks equally, their ability to make the most out of the anticipated acceleration in economic activity is likely to be influenced by initial conditions. It is in banks’ own interest to take action towards NPL resolution, to avoid that supply constraints induced by lingering bad loans become binding amid a sustained pick-up in credit activity. This is especially important in countries where the risks stemming from adverse interactions between weak banks and weak public finance are more pronounced. Last but not least, progress on this front will create favourable conditions for further advancement towards the institutional completion of the banking union, as advocated by the “Five Presidents’ Report”.
On its side, the ECB will continue, within its mandate, to use all its supervisory powers to help banks foster further progress towards risk reduction on their balance sheets.
Thank you for your attention.
 I am grateful to Francisco Ramon-Ballester for a first draft of this speech, and to Roberto Ugena Torrejon, Marco Rocco and Giordano Marchesin for helpful input and suggestions. I am solely responsible for the views expressed here.
 “Ethics in finance: a banking supervisory perspective”; remarks at the conference on “The New Financial Regulatory System: Challenges and Consequences for the Financial Sector”, Venice, 26 September 2014.
 See “ECB Guide on options and discretions available in Union law”, March 2016
 See “March 2017 ECB staff macroeconomic projections for the euro area”, available at https://www.ecb.europa.eu/pub/projections/html/index.en.html. See also “Monetary policy and the economic recovery in the euro area”, speech by Mario Draghi, President of the ECB, at The ECB and Its Watchers XVIII Conference, Frankfurt am Main, 6 April 2017, available at https://www.ecb.europa.eu/press/key/date/2017/html/index.en.html and the EC’s Spring 2017 Economic Forecast, available at https://ec.europa.eu/info/business-economy-euro/economic-performance-and-forecasts/economic-forecasts_en
 See “The Five Presidents’ Report: Completing Europe’s Economic and Monetary Union”, available at https://ec.europa.eu/commission/publications/five-presidents-report-completing-europes-economic-and-monetary-union_en