The SSM and international supervisory cooperation
Remarks by Ignazio Angeloni, Member of the Supervisory Board of the European Central Bank,
at the Symposium on “Building the Financial System of the 21st Century: an Agenda for Europe and the US”,
Eltville, 16 April 2015
I am very happy to be here this evening and to address this conference. And I am very grateful to the organisers for their flexibility in rearranging the programme so that I could participate.
The programme of this conference has attracted me from the outset. The agenda is broad, covering transatlantic issues. This fits with the current times, in which important regulatory changes are taking place in both the United States and Europe. Moreover, the themes cater for a mixture of economic, financial and legal expertise, with Harvard Law School and CEPS featuring as organisers, not to mention the Deutsche Bundesbank. I myself have an economics background, blended with banking and finance. But in recent years, I have increasingly come into contact with colleagues on the legal side, and I have acquired some of their mindset. Effective banking regulation and supervision require, evidently, a good balance among the three disciplines.
As you know, the Single Supervisory Mechanism (SSM) formally opened for business last November, after intense preparations lasting over two years. My best chance of saying something useful this evening is to focus on my experience in helping to establish the SSM. But the agenda of this conference is so rich that finding an original perspective was not easy. I have decided to focus on a particular component of the supervisory universe, that of international cooperation. Although the SSM is a new actor on the international cooperation stage, it is an important one, and not only geographically. Its potential contribution to improving the supervisory cooperation landscape raises legitimate hopes, but also reawakens some old concerns over the possible threat of a “fortress Europe”. In this regard, my conclusions are summarised in two statements:
Those fears should be dismissed. The opportunities that the SSM creates to improve the – indeed somewhat dysfunctional – state of international financial cooperation far outweigh those risks.
Whether the opportunities can be reaped quickly depends on how rapid the transition to the new supervisory environment is. The SSM needs to establish as quickly as possible its position on the global cooperation stage. For that to happen, some inconsistencies typical of the transition need to be overcome.
I will develop my argument in three parts: first, briefly mentioning the origins and rationales for supervisory cooperation (and, indeed, policy cooperation in general), then seeing how the inherited supervisory cooperation arrangements have evolved recently, and finally focusing on the potential contributions of the SSM.
Origins of supervisory cooperation
More often than not, in economic policy, new arrangements follow the events they are supposed to deal with, rather than preceding them as they should. Accordingly, over the recent decades international supervisory cooperation has strengthened in reaction to repeated turmoil in the financial markets, and also as a result of financial internationalisation. 
Three major milestones have marked the process.
The first happened more than 40 years ago, after the breakdown of Bretton Woods. In the exchange rate disorder that followed, many banks incurred losses. One of them – Herstatt Bank, a mid-sized bank located not so far from here – would become paradigmatic of a certain type of risk. At the time of the crisis, Herstatt’s foreign exchange settlement exposure amounted to three times its capital. Its banking licence was withdrawn by the German authority. Other banks which had unsettled trades with Herstatt also incurred losses.
That was a defining moment for international supervisory cooperation because it led, in 1974, to what would later become the Basel Committee on Banking Supervision. The Basel Committee is, still today, the main forum for global bank supervisory cooperation, having given rise, most notably, to the well-known accords on capital standards. In 1975, the notion of sharing supervisory responsibilities at the international level was enshrined in a paper produced by the Committee, the “Concordat”. In Europe, even earlier (in 1972), the so-called Groupe de Contact had promoted forms of collaboration and information exchanges. This committee was, by the way, the first one in which individual bank cases were discussed confidentially.
A second milestone, in the European Union, was the Second Banking Coordination Directive.  Its implementation led to discussions on establishing cooperation between the “home” and “host” supervisory authorities, resulting in around 78 bilateral Memoranda of Understanding being signed by the end of 1997.
A third event was the introduction of the euro in 1999, as a result of which financial markets became more integrated and interconnected, as documented by the ECB’s financial integration reports from that period. For the EU financial sector to remain functional, the regulatory framework also had to change. The Lamfalussy structure initiated the process with its three committees, subsequently enhanced and transformed into supervisory authorities in 2011.
In Europe, bank supervisory cooperation and convergence was strengthened, mainly by promoting the activity of colleges of supervisors for cross-border banks and by fostering convergence of supervisory practices. Rules for multilateral cooperation and coordination were laid down and signed by colleges of supervisors. This was an important step forward and set a precedent for further advances to come.
Benefits of supervisory cooperation
Economists have long struggled to measure the gains from economic policy coordination. A body of literature emerged in the 1970s and 1980s, when, following the demise of the Bretton Woods system, the world was looking for a new paradigm to re-establish order in international monetary and financial relations. Economic logic suggests that coordinating is beneficial because economic policy effects spill over across borders, but empirical analyses have typically found that such gains are quantitatively modest.  Most of these analyses, however, were conducted without properly incorporating banking and financial cross-border interconnections.
Several considerations suggest that the gains from coordinating regulatory and supervisory policies are probably greater than traditionally found.
First, given the growing cross-border nature of banks’ activities,  supervision with an exclusive national focus is insufficient because it does not internalise the cross-border transmission of risk. The result is likely to bias supervision towards laxity. Risk spillovers do not respect national borders, so the reach of supervisors should also not stop there.
The problem is compounded by the tendency of large multinational banking groups to conduct international business through establishments abroad – subsidiaries and branches. In 2008, before the crisis resulted in a retrenchment of international activities, the world’s five largest banking groups, which all have their headquarters in Europe and account for 10% of global banking assets, had over 50% of their credit risk exposures and approximately 60% of their employees outside of their home country.  In these conditions, effective supervision, either in normal times or in a crisis, cannot take place without adequate cooperation between “home” and “host” authorities.
From a bank’s perspective, poor supervisory cooperation unnecessarily raises the reporting and compliance burden and creates an un-level playing field. In crisis situations, weak and uncoordinated resolution practices induce countries to take unilateral actions in crisis situations, resulting in disorderly resolution and higher costs for taxpayers. 
Pitfalls of the pre-crisis framework
The pre-crisis supervisory cooperation framework appeared weak in a number of respects. 
First, geographical segmentation. Bank regulatory arrangements, including supervision and crisis management, were organised exclusively along national lines, with mild and non-binding cooperation frameworks at both the international and the European level. In several cases, such frameworks were also weak and incomplete within national borders.
Second, sectorial segmentation. In spite of most of the sector globally being organised along the universal banking model, regulatory practices followed a demarcation along sectorial lines (retail and investment banking, asset management, insurance), with weak or no established practices to exchange information or coordinate.
Third, no effective burden-sharing arrangements, especially when cross-border entities were involved. Cross-border negotiations on the allocation of losses during crises are difficult if no prior agreements exist. With bank failure costs borne domestically, via deposit guarantee schemes or taxpayers, the tendency is to ring-fence capital and liquidity buffers. Once this mechanism is activated, the likely outcome is an adverse loop of self-protection and market segmentation.
More generally, a contributing factor was the predominance of a consensus-based culture. Consensus decision-making is still the rule in international financial cooperation – in recent times Europe has started to become an exception in this regard. Unanimity tends to lead to inaction and give rise to minimal cooperation arrangements. This conflicts with the increasingly mobile and interconnected nature of the banking business.
Post-crisis responses: global and European
Interestingly, international cooperation among central banks has increased dramatically, and immediately, after the recent crisis. After the collapse of Lehman Brothers, in October 2008, the world’s major central banks announced that they were simultaneously lowering their policy rates. Foreign currency liquidity swaps by major central banks, including the ECB, were activated to deal with currency mismatches on the balance sheets of cross-border banks. The communication of this concerted action was unprecedented and was an important signal to market participants, that helped diffuse tensions. Cooperation has remained high since then.
In financial regulation, the response was less operational but more institutional. The most prominent examples, at the global level, were the reform of the G20 process and the transformation of the Financial Stability Forum into the Financial Stability Board (FSB), with the inclusion of emerging countries as members. Supported by an equally extended Basel Committee, an ambitious programme of financial reform was started. This is still underway and some important results have already been achieved. Going into details about this reform is beyond the scope of my remarks now. 
In Europe, emphasis was placed on institution-building, and, at a more operational level, on the functioning of colleges of supervisors as the mechanism to ensure effective supervision of cross-border banking groups. The three Lamfalussy committees were transformed into authorities, with power to issue standards, guidelines and recommendations, as well as to settle disagreements between authorities. The European Systemic Risk Board was created to provide analyses and recommendations in the field of macro-prudential supervision, also beyond the banking sector.
Moreover, Europe continued to play an important role in the development of colleges of supervisors. By the end of 2009, 33 colleges had been established, of which 30 had signed the written agreement for multilateral cooperation and coordination. The legal basis for colleges was strengthened in CRD IV, also with the involvement of third country authorities. The number of colleges steadily increased, amounting to approximately 100 colleges of supervisors in Europe prior to the establishment of the SSM.
This said, there are clear limitations on what colleges of supervisors can achieve. They are vehicles for coordination, not independent decision-makers. Colleges of supervisors can never fulfil such role without national intervention powers.
A second area where progress has been made is crisis management. Following the development in 2011 of the FSB’s Key Attributes of Effective Resolution Regimes for Financial Institutions, the European resolution framework was reformed with the Bank Recovery and Resolution Directive (BRRD). The BRRD provides for the creation of national resolution authorities in all EU Member States and for the application of harmonised rules for resolution, including for burden sharing. Also multilateral cooperation agreements addressing crisis events have been developed. Finally, a major step was the creation of a European authority for bank resolution, the Single Resolution Mechanism, that will start operating next year.
The role of the SSM
Let me now turn now to the Single Supervisory Mechanism, and the contribution it can make to supervisory cooperation.
A key starting point is the governance structure. The governing body of the SSM is its Supervisory Board, which is responsible for preparing all decisions. The Board has a highly cooperative structure. It includes six members that belong to the ECB: namely, the Chair, the Vice-Chair, and four representatives of the ECB. The latter have the function of providing a critical mass of voice and decision-making power in support of the singleness of the system and its European orientation. In addition, the Board includes a voting member for each national supervisory authority, normally the head or a responsible Board member of that authority. The comparative advantage of the national members is to provide detailed domestic expertise and perspective, which is essential in a wide and diversified membership; in addition, they retain responsibility, under the overall control of the ECB, for daily supervision of the less significant banks, constituting a large part of the national banking sectors. The national members are also bound by statute to act in the interest of the EU as a whole. The structure of the Board is highly cooperative, as I have mentioned, because it implies that each issue is discussed or deliberated with the constant and active participation of all authorities involved. I don’t think that, in the past, so many supervisors have ever conducted such an intense and regular consultation process as the one taking place now in the SSM Board.
In addition, the SSM has already made decisive steps forward in two directions, which I would call organisational singleness and operational singleness.
The organisational singleness concerns, first and foremost, the network of home-host relations, and the activity of the related supervisory colleges. This is, as I have already noted, an important issue in supervisory cooperation, linked to the fact that the banking industry is increasingly composed of cross-border entities. Looking only at the banking groups incorporated in the 19 countries of the SSM area, there are 190 subsidiaries in other SSM countries, plus 192 located in non-participating EU Member States and 692 situated in countries outside the EU.
Within the SSM, the structure of home-host supervisory has dramatically changed. Within the SSM area, the concept of home and host authority has disappeared. The colleges of supervisors that had been established for cross-border banking groups active only in SSM countries have been replaced by Joint Supervisory Teams (JSTs), groups of supervisors coordinated by the ECB, each responsible for the daily supervision of a significant banking group. In addition, for non-EU banking groups with sub-consolidated entities established within the SSM area, the colleges set up at the SSM level have also ceased to exist and have been replaced by the SSM representative as host supervisor. The difference between a “college of supervisors” and a “JST” should be emphasised: the colleges are only vehicles of coordination, whereas the JSTs are operational structures of the single supervisor.
The SSM has also drastically simplified the supervision of cross-border establishments in the EU. Non-SSM EU banks that wish to establish a presence in the SSM area and are deemed significant have to deal with only one host authority, the ECB. And the ECB is the sole home authority when it comes to the activities of significant banking institutions in the SSM area that wish to establish subsidiaries in non-participating Member States. In total, the SSM has become the single home supervisor for 32 colleges and takes part in another 16 colleges as host supervisor.
Similarly, the SSM allows a major simplification of cross-border establishments in the SSM area by banks incorporated in third countries.
Let me now turn to the operational singleness. To achieve it, the SSM relies upon three main building blocks.
The first is a comprehensive Supervisory Manual, describing its supervisory processes, methodologies, and cooperation mechanisms within and outside of the EU. The Manual builds on best practices from national experience, merged into a coherent whole. It guides our direct supervision over the significant institutions. The Manual also includes methodologies on how NCAs should conduct the supervision of less significant institutions.
The second building block concerns the structure and activity of the JSTs. These teams are composed of staff from both the ECB and the NCAs, working under an ECB coordinator. This set-up combines the expertise and vision of the ECB with the specific experience and knowledge of the national supervisors, providing a degree of peer control and a valuable combination of different supervisory cultures.
The third building block relates to the internal organisation of the ECB supervisory staff, which includes specialised expertise at all levels in the supervisory process. This work is conducted together with networks of national experts, contributing to the operational singleness of the SSM
Let me now briefly touch upon two areas where work is underway but progress is still incomplete.
The first relates to the cross-border cooperation agreements among supervisory authorities. Traditionally cross-border cooperation has been codified, on a purely voluntary basis, in Memoranda of Understanding, or MoUs. Over the years, MoUs have piled up on a variety of issues (information exchanges, coordinated interventions, crisis management, etc.), typically on a bilateral basis, with little or no overall vision or control.  It is difficult even to understand how many of these MoUs exist, let alone make overall sense of them. Evidently, MoUs agreed between authorities that are now in the SSM have lost their purpose: a welcome simplification in itself. But there remain, according to our tentative counting, around 40 MoUs between SSM and non-SSM EU authorities, as well as around 170 MoUs between authorities of the SSM and third countries. 
As a first step, and so as not to disrupt existing relations, the ECB has asked to take over the participation in existing MoUs between the SSM countries and over 80 third countries. As you can imagine, compiling a census of this “universe” requires a lot of work: mapping the existence of agreements, analysing their contents and developing contacts with the counterparties. Our next step is to define the ECB’s own cooperation agreements, as coherently as possible across counterparties. To this aim, the ECB is working on a standard template to be negotiated with non-SSM partners, leveraging on national experience, but reflecting the new European imprint of the SSM.
I have left as my last point, before concluding, an issue that is perhaps the thorniest of all: that of the international representation of the SSM.
A supervisor from another planet, landing on earth to study our banking union, would probably assume that the SSM had taken over the representation in the relevant cooperation fora, having taken over the operational responsibility of supervision. We are very far from that logical endpoint. Anyone familiar with the difficulties of reforming the representation structure of the IMF Board understands that it may take years, if not decades, before the SSM occupies in those fora a position that matches its effective responsibility as supervisor.
Some steps have been made. Since last September, the SSM has its own seat in the Basel Committee on Banking Supervision. It also participates in the Board of Supervisors of the EBA, but without voting rights. Membership of the SSM in the Financial Stability Board is being explored. In all of these fora, the supervisors of all main EU Member States hold a full seat with voting rights .
Avoiding inconsistency of positions during meetings with ensuing reputational damage vis-à-vis our international partners is often a challenge; much is left to the discretion and sensitivity of individual participants.
Let me conclude now. I hope that, with my short remarks, I have given you an idea of how complex and long the process to bring more order and effectiveness to international supervisory relations is. At the same time, I hope that I have convinced you that the SSM is a great novelty in this respect as well, capable of contributing towards simplicity, order and transparency.
I am convinced that progress is possible; it depends on how well and how hard we work on it. Progressing will be in the interest of Europe and of our international partners alike.
Thank you for your attention. I am now happy to take your questions.
I am grateful to Alexander Hodbod, Anton van der Kraaij and Cécile Meys for excellent drafting support, and to Gilles Noblet, Jakob Orthacker and Stefan Walter for useful comments. The views expressed here should not be attributed to the Supervisory Board of the ECB.
A good reference to some of the topics in this and the following sections is the book by H. Davies and D. Green, Global Financial Regulation: The Essential Guide, Polity Press, 2008.
Second Council Directive 89/646/EEC of 15 December 1989 on the coordination of laws, regulations and administrative provisions relating to the taking up and pursuit of the business of credit institutions and amending Directive 77/780/EEC (OJ L 386, 30.12.1989, p. 1).
A short survey of this literature is contained in I. Angeloni and J. Pisani-Ferry, “The G20: characters in search of an author”, Bruegel Working Paper Series, No 4, 2012.
Ample evidence of this, for the years before the crisis, is provided by the statistics of the Bank for International Settlements; McGuire, P. and Tarashev, N., “Global monitoring with the BIS international banking statistics”, BIS Working Paper Series, No 244, January 2008 ( www.bis.org/publ/work244.pdf).
IMF, “Cross-border bank resolution: recent developments”, IMF Policy Paper Series, June 2014, p. 5 ( http://www.imf.org/external/np/pp/eng/2014/060214.pdf).
Maurer, H. and Grussenmeyer, P., “Financial assistance measures in the euro area from 2008 to2013: statistical framework and fiscal impact”, Statistics Paper Series, No 7, ECB, Frankfurt am Main, April 2015. Financial needs to support the financial sector from 2008 to 2013, as a percentage of GDP: e.g. Germany 8.8%, Ireland 37.3%, Italy 0.2%, France 0.0%, Spain 4.9%, euro area 5.1%.
For further details about those arrangements, see I. Angeloni, “Testing times for global financial governance”, Bruegel Essay and Lecture Series, 2008.
For an early summary see I. Angeloni, “The Group of 20: Trials of global governance in times of crisis, in B. Eichengreen and Bokyeong Park (eds) (2012) The global economy after the financial crisis, World Scientific Publishers. A more recent and critical account is provided by N. Véron, “The G20 financial reform agenda after five years”, Bruegel Policy Contribution Series, No 11, September 2014.
A landmark was the 2008 Memorandum of Understanding on cooperation between financial supervisory authorities, central banks and finance ministries of the European Union on cross-border financial stability. It was signed by the 114 authorities with competence on resolution in the 27 EU Member States. The text was made public to signal the EU’s preparedness for potential banking crises. Unfortunately, it did not prove very helpful during the recent crisis.
These only refer to MoUs dealing with general issues of cooperation, and do not include those that deal with the supervision of specific banking groups.