On the road to a banking union: the way forward from the ECB’s perspective
Speech by Yves Mersch, Member of the Executive Board of the ECB,
at the 16th Euro Finance Week,
Frankfurt am Main, 18 November 2013
Ladies and gentlemen,
Last year, questions surrounding the banking union concerned fundamental concepts: How can we break the close links between countries and their banks? How can Europe’s fragmented financial market become better integrated again?
While the objectives have remained the same, the public debate is now focusing more on the details of its implementation: What exactly will the balance sheet analysis look like? And how will balance sheet assessments and stress tests be combined? How, precisely, will the ECB’s supervisors cooperate with their colleagues from the national supervisory authorities?
To the extent possible, I would like to begin today by answering some of these questions. I will also return to the key conceptual idea behind the banking union: the desire to establish uniform basic conditions for everyone all over Europe – conditions that will allow us to break the close links between countries and their banks and establish a single financial market.
Specifically, I will address two questions:
First, how is the fundamental idea behind the banking union reflected in the practical design of the single system of banking supervision?
Second, what else can we do to be faithful to that fundamental idea when banks cease to be viable and have to be wound up?
Let us look, first of all, at the fundamental idea of ensuring uniform conditions. If we are to have an effective European policy, there are decisions that have to be taken when it comes to implementation. We can have rules that go beyond mere harmonisation – very rigid rules that do not allow for any exceptions. That will mean that implementation decisions and instruments can continue to be taken and employed in a decentralised manner.
Or, alternatively, we can have a set of rules that contains a certain amount of room for discretion, one that takes account of the particular circumstances of individual cases. That, however, will necessitate a central body to apply and implement those rules. Exceptions must be objectively justified and cannot reflect the special interests of individual countries. That is the only way to guarantee fair and consistent basic conditions across Europe.
The same is true of the banking union. Uniform conditions across Europe are aimed at establishing a single banking system to accompany the single currency.
This is not a case of “more Europe” just for the sake of it. This issue is addressed in the Treaty on European Union under the heading “subsidiarity”. Decisions should only be taken at European level if action is more effective at that level than it would be at national level.
What does this mean in practice?
I will begin with the comprehensive assessment of the major banks of the euro area, which will be the focus of our new supervisory function. This may well concern many of the people here in this room.
This stocktaking exercise will comprise three elements: a risk assessment; the balance sheet assessment itself (analysis of the quality of assets); and a stress test, which we will conduct in close cooperation with the European Banking Authority (EBA).
The risk assessment seeks to determine the riskiest asset classes and portfolios. Factors such as refinancing and liquidity risks are examined.
Building on the risk analysis, which has already begun, the balance sheet assessment itself is scheduled for the first half of next year, after the annual financial statements for 2013 have been drawn up. We will then look in detail at a selection of individual assets in the portfolios identified. We will also be looking, in that regard, for appropriate risk provisions at banks. Essentially, it is a case of ensuring that there are no unknown risks hiding in banks’ balance sheets – for example as a result of inappropriate classification or valuation of assets.
That will give us an overview of whether the banks have a viable business model and where the biggest risks lie.
The balance sheet analysis itself will be conducted on the basis of uniform criteria in three stages:
Portfolios will be selected on the basis of proposals by national supervisory authorities, which will be checked – and potentially amended – by the ECB.
In the course of this process, data will be checked for integrity and samples will be selected. Checks will then be carried out to determine the extent to which the valuation of the various assets and the classification of non-performing loans, collateral and provisions are appropriate.
Finally, the results will be compiled, collated and checked for consistency.
The banks of the euro area are certainly in a significantly better position than they were at the beginning of the crisis. They have taken on €225 billion of fresh capital. Governments have also made available additional capital injections totalling €275 billion. However, it may be that our investigations uncover shortfalls in terms of capital.
The question, then, is how those shortfalls can be remedied. Initially, the banks concerned must use the market to cover their capital needs. However, in the event that individual banks are unable to raise the necessary capital, we also require robust alternative solutions to provide them with support.
While the balance sheet analysis is a static assessment of the current situation, the stress test seeks to offer a dynamic diagnosis. We consider that conducting this test over a period of three years will be enough. It should be sufficient, in that regard, for us to use a baseline scenario and one stress scenario.
For the balance sheet assessment and the baseline scenario in the stress test, we foresee a core capital ratio (common equity; tier one) of at least 8%. This is in line with the requirements for 2014 in the implementation rules for Basel III.
A further issue, which we are currently discussing internally, concerns the question of how exposure to government bonds is to be valued. While in order to comply with current legislation, a capital requirement of 0% is necessary for government bonds in the balance sheet analysis and a distinction has to be drawn between the banking book and the trading book, no decision has yet been taken on the requirement to be imposed for government bonds in the stress test. But solely by taking account of the market risk and the duration of the stress test combined with the maturities of the bonds, these assets inevitably will be stressed.
Sometimes, the desire for full harmonisation and efforts to increase efficiency is limited by what is feasible. For example, the EBA has produced a comprehensive common definition of “non-performing loans”. However, in order for it to acquire legal force, this definition must first obtain the blessing of the Commission. Although we use this definition, some banks will not be able to adjust their internal systems in time. For that reason, a simplified variant is recognised as a minimum standard. In any event, this still goes further than the current chaotic mass of national provisions and lays foundations in this regard at European level.
The national supervisory authorities are responsible for making the necessary data available. Thus, we respect the principle of subsidiarity. But they do so on the basis of very clear, uniform guidelines. And the final quality controls and plausibility checks are conducted centrally at the ECB. This allows us to ensure, again, that the same conditions apply everywhere.
However, this comprehensive assessment is merely preparation for the actual supervision, which we will begin in one year’s time. The same principles of uniform basic conditions and maximum use of national expertise will apply in the day-to-day work of Europe’s banking supervisors.
The ECB will directly supervise the most important banks in the euro area. In order to ensure uniform conditions and respect for the principle of subsidiarity, we will be setting up joint supervisory teams. This will enable us to harmonise supervision centrally and make use of the expertise of the national supervisory authorities.
Here is an example to help explain what exactly “joint supervisory teams” means. Each major bank will be supervised by a team of ECB supervisors and national supervisors, led by a central ECB coordinator. For each euro area country where this major bank is active, there will be a sub‑coordinator from the relevant national supervisory authority. The ECB coordinator, who has primary responsibility, will be tasked with carrying out the final quality check. He/she will also be responsible for reporting to the Supervisory Board.
The fact that the ECB is directly supervising only the most important banks does not mean that all other institutions will remain subject to national requirements. All banks will be subject to the same rules and procedures when it comes to supervision. For example, we have drawn up harmonised templates for the reporting of data, and all on-site assessments are subject to the same rules.
Implementation remains the responsibility of the national supervisory authorities, but the ECB, as an established supranational institution, has the final say. Moreover, if the ECB considers it appropriate, it can take direct control of the supervision of a bank at any time. Both of these measures ensure that uniform criteria do not just exist on paper and are indeed applied in practice.
In short, the single supervisory mechanism ensures a harmonised system of banking supervision. At the same time, supervision of smaller institutions continues to be carried out on a decentralised basis. This enables us to ensure that everyone is subject to the same conditions – fully in keeping with the further integration of Europe’s Single Market – while at the same time using the expertise of national supervisors.
This combination of national supervisory authorities and the ECB at a supranational level has at least three advantages.
First, harmonised technical standards reduce costs for banks. They also make it easier to compare banks in different countries.
Second, it is easier for national supervisory authorities to coordinate their actions when it comes to supervising banks that are active in more than one country. This makes the whole process significantly more efficient than the current colleges of supervisors, which will remain in place only for banks not covered by the new regime, and will be managed by the ECB.
Third, the ECB, as an independent supranational institution, will ensure that banking supervision ceases to be influenced by national interests. Instead, it will be conducted in the interests of Europe as a whole.
However, supranational banking supervision alone will not be enough to overcome the fragmentation of Europe’s financial market. We also need suitably European solutions as regards the resolution of banks.
There are very close links between the countries of Europe and their banks. The credit default swap (CDS) spreads of banks and countries in the euro area are almost perfectly correlated. In the period between the beginning of 2010 and August of this year, the correlation coefficient between the average CDS spreads of large and complex banking groups in the euro area and the CDS spreads of their home countries was 0.9. In the United States, on the other hand, it was just 0.2.
These close links between countries and their banks are to be loosened using harmonised European rules on recovery and resolution. Ultimately, this always comes down to the same question: Who pays when a bank gets into difficulty?
The planned directive harmonising the recovery and resolution of banks (the BRRD) provides for a clear hierarchy of liability in this regard. It first falls to the banks’ owners, the shareholders, and then the creditors. The government can only get involved as a very last resort.
In my view, however, these rules still allow too much room for manoeuvre at national level. And that brings me back to the fundamental concept. A certain amount of room for manoeuvre within a harmonised set of rules can be entirely appropriate from the perspective of financial stability. In that case, though, those rules must be applied at a supranational level if the same conditions are indeed to apply everywhere. If decisions are taken by national decision-making bodies, the rules have to be so rigid that we end up with a uniform set of European conditions.
Problems arise when the room for manoeuvre in the set of rules means that the envisaged involvement of shareholders and creditors is fully implemented only in countries that experience extreme financial difficulties. If creditors believe that a bank from a country with relatively healthy public finances will be saved using taxpayers’ money in the event of an emergency, but that they themselves will have to pay in the case of banks from crisis countries, they are hardly likely to accept the same rates of interest. That then leads, in turn, to increased fragmentation of the financial market, as banks with that kind of implicit government guarantee have significantly lower refinancing costs. According to a study conducted by the OECD, this implicit guarantee allows the 17 largest German banks to save more than €20 billion in interest every year. 
Once the single bank resolution mechanism is in place, the planned resolution authority will apply the rules everywhere in a consistent manner. A certain degree of discretion will then certainly be justified in the interests of financial stability, as it will be accompanied by a central implementing authority.
That single resolution authority will be tasked with ensuring two things: first, that banks that have ceased to be viable are wound up at as little cost as possible; and second, that such costs are borne by those who caused them – not taxpayers.
At this point, it is worth taking a quick look across the Atlantic. The US equivalent of this single resolution authority is the Federal Deposit Insurance Corporation (FDIC). Since 2008 the FDIC has wound up around 490 banks, without causing major turmoil in the financial market. What is more, taxpayers have also been largely spared. We in Europe can certainly learn from that.
In 450 of those bank resolutions, the FDIC has used solutions such as mergers and acquisitions. That approach has proved to be a success. The fact that parts of struggling banks have been bought by other banks has meant that the costs caused by failing banks have largely been borne by the banking sector itself, rather than having to be shouldered by taxpayers.
In Europe, on the other hand, governments have tended to intervene when banks have threatened to collapse. That is one of the reasons why many of those countries have themselves ultimately ended up seeking help in the form of an EU/IMF programme.
The FDIC’s strategy, however, has helped to establish a better integrated and more robust banking sector.
Thus far, though, the idea of using cross-border acquisitions as a recovery strategy has been scarcely thinkable in the euro area. First, many banks are unwilling to acquire parts of other banks abroad. And second, such a strategy would require close coordination between national resolution authorities.
This kind of coordination has proved difficult thus far. For that reason, I am also sceptical when it comes to the idea of a loose national network of resolution funds within the framework of a single resolution mechanism. In my view, this is another area where a clear division of responsibilities and a European solution are required.
I therefore welcome the European Commission’s proposal for a single resolution mechanism, as it envisages a single European system, a single authority and a single fund.
We now need to turn this proposal into a functioning resolution mechanism by the target date of the beginning of 2015. For we in Europe urgently require a full banking union with both fundamental elements: harmonised supervision and harmonised resolution. As the fund is to be financed by the banks, which will take some time to arrange, the primary question concerns the issue of transitional solutions. These, in turn, raise questions relating to the division of responsibilities. Mistakes of the past at national level should not, in principle, be paid for by a joint fund. However, if there is no willingness to establish a stopgap measure, the credibility of this new architecture strengthening the single currency will be questioned.
I would now like to conclude by summing up. European policies will only be successful if we strike the right balance between room for discretion and centralisation. At the same time, the principle of subsidiarity needs to be respected.
We are taking account of both in establishing the single banking supervision system. And the same should be true of the design of the single bank resolution mechanism. A few details still need to be agreed in that regard.
I believe that this can function well, if we again remember the fundamental concepts, for I am sure that we are all now in agreement on those. I believe that we all want a functioning European financial market, as well as banks that are properly supervised and can be wound up if necessary, without threatening financial stability or imposing unnecessarily large costs on taxpayers.
Those principles should guide all further work on the design of the banking union.
The objective, then, is clear. Now we just have to ensure that we achieve it in a timely manner. To conclude with the words of a famous Frankfurt poet:
“A good many men have an idea of the goal, only they would like to reach it by sauntering along the labyrinthine ways.” (Johann Wolfgang von Goethe).
Schich, Sebastian and Lindh, Sofia (2012), “Implicit Guarantees for Bank Debt: Where Do We Stand?”, OECD Journal: Financial Market Trends, Volume 2012, Issue 1.