Another look at proportionality in banking supervision
Remarks by Ignazio Angeloni, Member of the Supervisory Board of the ECB, at the Thirteenth Asia-Pacific High Level Meeting on Banking Supervision, Singapore, 28 February 2018
It is a pleasure to be here today. I am grateful to the organisers for inviting me and in particular for asking me to speak in this panel on proportionality in banking supervision, a topic which, I shall argue, appears less straightforward after reflection than it does at first sight. 
The notion of proportionality is so deeply rooted in our culture, and so engrained in our common understanding, to seem to require no further thought. The word comes from the Latin pro portione – the correspondence of measures between two or more related concepts. The Roman architect Vitruvius elaborated on the ideal proportions in architecture, relating them to the human body. This inspired the famous drawing by Leonardo da Vinci (the “Vitruvian Man”), which depicts a man of ideal proportions inscribed in a circle and a square; a universally recognised icon of harmony and beauty which, incidentally, also appears on the national side of one of our euro coins. But even keeping all this aside, who would deny that proportionality is to be applied always and everywhere? Would anybody ever consider advocating dis-proportionality? For sure not.
Yet, as one moves from concepts to concrete applications in banking supervision, pitfalls begin to appear and a good deal of care is needed. Proportionality means adapting nature and intensity of supervision to the specifics of the bank – its size, its risk profile, its business model – and to the particular purpose being achieved. In practice this means adopting, or coming close to, a case-by-case approach. In doing so, proportionality meets the boundary with two other principles: prudential soundness and competitive level-playing-field. The first is the number one priority of banking supervision. The second, relevant everywhere, is especially important in the euro area given its supra-national dimension.
In my remaining remarks I want to briefly address three issues: first, how is the principle of proportionality translated in our legal and regulatory framework; second, what does European Banking Supervision do to conform to that principle; third and final, what are the challenges one faces and the limits one must respect in applying proportionality in supervision.
The legal basis and the ECB experience
I don’t need to remind this audience that proportionality features prominently in the Basel Committee framework, being part of its Core Principles. Principle 8 calls on supervisors to “develop and maintain a forward-looking assessment of the risk profile of individual banks and banking groups, proportionate to their systemic importance”. Recently, the Financial Stability Institute of the BIS has issued a paper which provides useful international comparisons of how proportionality is applied by supervisors around the globe.
Proportionality is well-established in the European Union (EU) Treaty. Article 5(4) notes that “under the principle of proportionality, the content and form of Union action shall not exceed what is necessary to achieve the objectives of the Treaties. The institutions of the Union shall apply the principle of proportionality as laid down in the Protocol”. Proportionality thus applies to all EU acts and EU institutions and may be used to challenge Union action itself.
Coming to EU banking law, Recital 46 of the Capital Requirements Regulation (CRR) notes that “the provisions [of this Regulation] respect the principle of proportionality, having regard in particular to the diversity in size and scale of operations and to the range of activities of institutions”. The principle of proportionality is explicitly mentioned in several areas, for example in relation to reporting (CRR Article 99), internal capital adequacy assessment plans (Capital Requirements Directive IV, or CRD IV, Article 73), and recovery and resolution plans (CRD Article 74). Recital 55 of the Regulation which confers supervisory duties to the ECB (known as the ‘SSM Regulation’) states that the central bank should “use its supervisory powers in the most effective and proportionate way”. Proportionality is thus one of the key principles governing the functioning of European Banking Supervision, with supervisory intensity varying according to a bank’s size and complexity.
The principle of proportionality is enshrined in the organisational structure of the Single Supervisory mechanism (SSM), the system of banking supervision which we have established in Europe comprising of both the ECB and the national supervisory authorities of the participating countries. The ECB directly supervises some 120 significant banking groups which together account for around 900 individual entities, holding a combined 82% of euro area banking assets. For all other euro area banks (around 3,200) individual supervisory decisions are made by National Competent Authorities (NCAs), with the ECB performing an indirect oversight function over their actions. This distinction is an operational necessity: a proportionate way to ensure an efficient supervision over such a large and diversified bank population must rely on a division of labour between the ECB and the national authorities, with the latter focusing mainly on banks that are less significant within the area-wide banking sector. Significance is measured on a set of criteria that include size, importance to the domestic economy, and extent of cross-border activities.
To avoid any misunderstandings, I should state once more that the distinction of banks in these two categories does not imply differences in the supervisory approach. The SSM is not, and should never become, a “two tier” system where different supervisory styles apply and different risk tolerance levels are accepted. Direct and indirect supervision are different ways of achieving the same purpose. In fact, one responsibility the SSM Regulation assigns to the ECB is precisely to ensure the uniqueness of the system. To achieve that, the ECB may assume the direct supervision also of less significant banks, if this is needed to guarantee a “consistent application of high supervisory standards”.
Let me now mention other areas where proportionality applies in the SSM.
The first relates to significant banks, where reporting requirements and fees already differ. Moreover, while a minimum level of supervisory engagement is maintained for all banks irrespective of their risk, the intensity of the supervisory action is based on a bank’s risk profile. All prudential measures decided on the basis of the ECB’s annual review of banks’ risks (the Supervisory Review and Evaluation Process or SREP), including Pillar 2 capital requirements, are tailored to the individual risk profile and characteristics of the bank in question. Relatedly, the size of the Joint Supervisory Teams, which are the functional units composed of both ECB and NCA staff responsible for the direct supervision of any given bank, is linked to the size and complexity of the entity.
The second area refers to less significant banks, where the scope and frequency of information requests for the purposes of reporting requirements depends on the nature of the entity concerned, and whose supervisory fees are also linked to the institution’s significance level.
The third area applies again to less significant banks. These are prioritised into High, Medium and Low priority entities according to a methodology which the ECB jointly developed with NCAs, based on the riskiness of the bank concerned and its potential impact on the domestic economy. Prioritisation influences the level of cooperation between the ECB and national authorities and the intensity of supervisory oversight. The collection of information and the allocation of supervisory resources take into account the priority level. Prioritisation also drives the SREP process for such banks, which has significant elements of proportionality embedded in it, including the frequency and the depth of the required analysis.
Let me mention here some initiatives under discussion at the international and European levels to enhance proportionality in banking supervision. Within the BCBS, simpler approaches for calculating capital requirements (e.g. credit risk, operational risk, CVA risk) have been developed. At the EU level, in the ongoing review of the CRR and CRD IV, the Commission has proposed simplified market and counterparty credit risk measures for smaller institutions as well as reductions in the frequency and scope of supervisory data reporting and disclosure. The ECB does not support lowering the reporting frequency for such entities. Frequent reporting is a key source of information for supervisors, especially in relation to smaller institutions which do not undergo sufficient market scrutiny. Early warning systems are effective only if data are sufficiently frequent. Another challenge stems from setting the thresholds to categorise institutions, because quantitative thresholds are difficult to apply and at the margin may create cliff effects and distortions.
Challenges and limits going forward
In recent times, we hear a growing demand for more recourse to the principle of proportionality in supervisory and regulatory practice. In the US, community banks have been a focus of supervisory concern at least since the mid of this decade. In Europe, the issue is particularly felt in countries where the presence of small banks is more important, first and foremost in Germany, Austria and Italy. I already mentioned the involvement of the BCBS. The European Banking Authority issued a report in 2016. Proportionality is attracting more interest in supervisory debates, as this panel also suggests.
There may be various reasons for this revival. After the financial crisis, the scope and intensity of banking regulation has increased; new and stringent rules were introduced, covering, beyond bank capital, liquidity, funding, leverage, governance and controls, individual fitness and propriety, resolvability, and so on. Not only the volume of regulation, but the complexity and burden of reporting and compliance have increased. In the EU, the degree of harmonisation has grown due to the transition of many provisions from the realm of Directives to that of Regulations, which are applied directly and equally across member states. In the area adhering to the banking union, the establishment of the SSM has marked a decisive move towards centralisation of supervision, in a geographical area characterised by a diverse banking population and by different supervisory traditions.
In this environment, proportionality is seen by advocates as a way to mitigate the burden on banks, primarily the small ones. Surveys for the US and Europe show that regulatory compliance costs are higher in relative terms for smaller banks relative to larger intermediaries, a rather straightforward finding since compliance involves fixed costs (e.g. relating to IT and reporting) whose weight varies inversely with size. One also gets the feeling that the demand for more proportionality is reinforced by a desire, in today’s improved economic and financial climate, to call an end to the crisis and swing the pendulum a bit back towards a less regulated and more business-friendly approach.
A fundamental question arising in this context is whether or not prudential frameworks should accord a more favourable treatment to small banks. Arguments in this regard are complex and go in different directions. Three lines of reasoning come to mind. The first one is based on riskiness; if small banks have simpler and safer business models, a more lenient supervisory approach may be justified. The second argues that the presence of many small entities improves the functioning of the market, either because of more competition, or because diversification reduces herd effects, or perhaps because small banks are associated, as it happens in other sectors, with a greater potential to innovate. A third and long-standing line of reasoning contends that small banks contain some “social” value, because they are capable of entertaining more stable and reliable relations with their clients – typically SMEs and households – especially in local markets.
Let’s briefly examine each of these arguments.
In judging the riskiness of a bank, the macro- and the micro-prudential dimensions must be distinguished. Smaller banks transmit little or no risk to the system, but this does not imply that they are individually less risky. Evidence for the euro area after the establishment of the SSM seems to suggest the opposite: for most of the period, the number of bank failures has been much higher in less significant than in significant banks, both in absolute terms (total number of cases) and in relative terms (in relation to the number of existing banks). The year 2017 is somewhat special, because of the almost coincident declaration of failure of three significant entities by the ECB. In the US, historical data provided by the Federal Deposit Insurance Corporation show that the bulk of bank failures take place in smaller banks, and failure rates are only marginally higher for larger banks once state support during crisis periods is factored in (with banks being considered as “failed” on this basis). All in all, while comparisons of bank failure rates across bank population segments need to be interpreted with great caution, the evidence does not seem to support the existence of systematic and large differences in failure rates across size classes, which may justify a different micro-prudential treatment. Prudential soundness must remain the overarching principle, regardless of size. The higher systemic relevance of larger banks should, of course, be addressed by higher macro-prudential charges.
The second line of reasoning may be more relevant, but qualifications are needed there as well. Small banks indeed promote diversification in the banking system, hence reducing concentration and correlation, if that they act independently from each other. This contributes to stability. However, especially in Europe small institutions are often part of larger groups or associations (such as institutional protection schemes) which reduces the benefit from diversification. The tendency to aggregate in groups or protection schemes clearly confirms the existence of powerful economies of scale and scope. Small banks may play a role in stimulating competition in some segments (retail deposit-taking and lending), but again, overcoming the scale disadvantages is hard and there seems to be no reason for the regulator to distort the level-playing field. There also seems to be little ground to favour small banking on grounds of innovation. Innovation in banking and finance comes mainly from large banks (which have broader customer base and investment capacity) or from outside the banking sector (e.g. non-banking groups or specialised Fintech firms).
Finally, a delicate judgement relates to the potential social role of small banks. We approach, here, the boundary between prudential and political considerations. Small and cooperative banks are still widely recognised as having a social role by granting access to finance, or the provision of other social benefits, particularly in certain areas and for specific population groups. In Europe especially, savings and cooperative banks maintain a strong historical and cultural rooting. On the other hand, it cannot be ignored that most of the cooperative sector has moved away from its original social and charitable function, and today performs standard banking business providing services undistinguishable from those of its competitors. There may be cases justifying a support to smaller banks which provide financial services where other banks are less present, but there seems to be no compelling reason why such support should be embedded in the prudential framework.
I would summarize my conclusions in four statements.
- Proportionality is a well-established principle of supervision and, indeed, of public policy more generally. In Europe, this principle is recognised by the Treaty and its application is governed by provisions in banking legislation and regulation.
- In the SSM, proportionality is built in its organisational structure and informs many of our daily supervisory practices.
- Further major advances in the application of proportionality, as now proposed by some in order to soften the prudential burden for certain categories of institutions, should be regarded with great caution. Proportionality should not lead to supervisory laxity or deviations from the single rule book. For example, the ECB does not favour a modification of the existing legal framework to allow for less frequent reporting requirements for smaller banks.
- Arguments and evidence do not support, on balance, a more favourable treatment to small banks by the micro-prudential regulator or supervisor. If deemed justified, for example for social reasons, such support should be provided otherwise. In the SSM, prudential soundness, harmonisation and competitive level-playing field have been and should remain the overarching guiding principles.
Thank you for your attention.
 I am grateful to Francisco Ramon-Ballester and Carlo Capra for preparing a first draft of this speech, and to Giuseppe Siani for helpful suggestions. I am solely responsible for the views expressed here and for any errors.
 See Castro Carvalho, A., S. Hohl, R. Raskopf and S. Ruhnau (2017), “Proportionality in banking regulation; a cross-country comparison”, FSI Policy Implementation Insight n. 1, Financial Stability Institute, available on the BIS website
 See Regulation (EU) No 575/2013 of the European Parliament and of the Council of 26 June 2013 on prudential requirements for credit institutions and investment firms and amending Regulation (EU) No 648/2012, available on the EUR-Lex website
 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 on the EUR-Lex website
 See “Some Thoughts on Community Banking: A Conversation with Chair Janet Yellen”, Community Banking Connections, second quarter 2015, available on the CBCFRS website. See also “A Conversation About Regulatory Relief and the Community Bank,”, remarks by FDIC Vice Chairman Thomas Hoenig at the 24th Annual Hyman P. Minsky Conference, National Press Club, Washington, D.C, April 2015, available on the FDIC website
 See “BVR report on the impact of regulation: Disproportionate strain on small and medium-sized banks”, report by the National Association of German Cooperative Banks, September 2015, available on the BVR website. See also three recent speeches by Andreas Dombret, Member of the Executive Board of the Deutsche Bundesbank: “Heading towards a ‘small banking box’ – which business model needs what kind of regulation?”, given at the Bavarian Savings Bank Conference, June 2017; “On the Road to Greater Regulatory Proportionality”, given at the strategy conference of the Rheinland Savings Banks and Giro Association, August 2017; and “Sometimes Small is Beautiful, and Less is More; a Small Banking Box in EU Banking Regulation”, given at the Representation of the State of Hessen to the European Union, Brussels, October 2017; all three available on the Bundesbank website
 The Italian Banking Association (ABI) is preparing a set of proposals addressed to European regulators.
 See the keynote speech by Bank of Greece Governor Stournaras at the Conference on “Proportionality in European Banking Regulation” organised by the Bank of Greece and the University of Piraeus, 2017
 For the US community bank sector see Dahl, D., A. Meyer and M. Neely (2016), “Scale Matters: Community Banks and Compliance Costs”, Federal Reserve Bank of St Louis. For Germany, see Schenkel, A. (2017), “Proportionality of Banking Regulation - Evidence from Germany”, University of Muenster, Institute for Cooperative Research.
 However, in Europe it is often the case that small banks are part of wider networks – such as institutional protection schemes – that allow to pool those services and to share their costs. Small banks typically have a simpler balance sheet and presumably a less risky business model.
 See “Is small beautiful? Supervision, regulation and the size of banks”, statement by Sabine Lautenschläger, Member of the Executive Board of the ECB and Vice-Chair of the Supervisory Board of the ECB, at an IMF seminar, Washington D.C., October 2017
 FDIC data show that there were 573 cases of bank failures (including 13 state support cases) over the 2000-2017 period, but only 10 of these banks had assets exceeding USD 30bn at the time of failure (or just 2 of these banks excluding state support). The picture does not significantly change if the size threshold is lowered to USD 10bn. Relative failure rates over the 2000-2017 period among banks with assets of USD 30bn or more expressed as a percentage of banks existing in 2017 drop markedly to 3.3% from 16.4% once state support is excluded. According to the same metric, failure rates among banks with less than USD 30bn assets at the time of failure were around 10% regardless of the presence of state support. If the size threshold is lowered to USD 10bn, relative failure rates excluding state support expressed as a percentage of banks existing in 2017 are still higher among small banks (9.8%) than large banks (7.1%). See the FDIC web page on failed banks
 The higher systemic relevance of large banks should, at least in principle, be addressed by the higher macro-prudential charges imposed on them.
 See “Savings banks in the European banking landscape”, speech by Ignazio Angeloni, Member of the Supervisory Board of the ECB, at the Financial Meeting of the Confederación Española de Cajas de Ahorros, Madrid, June 2014