Claudia Buch: The banking union - setting the course for better integrated and more stable financial markets in Europe?

Speech by Prof Claudia Buch, Deputy President of the Deutsche Bundesbank, at the 19th Handelsblatt Annual Conference, London, 4 September 2014.

The views expressed in this speech are those of the speaker and not the view of the BIS.

Ladies and gentlemen

I would like to express my sincere thanks for the invitation to speak to you today.

The topic of this conference is "banks in transition". It is not just the banking sector, but also banking supervision, that is experiencing a transitional phase.

The creation of the banking union is a mammoth project which is being implemented with breathtaking speed. In May 2013, a comment in a newspaper - not the Handelsblatt¹ - included the following line: "No self-respecting European is refraining from calling for the speedy launch of the banking union in public speeches at present." Today, just under 16 months later, the banking union is truly just around the corner. The question now is whether it can actually fulfil the expectations and hopes bound up in it.

My predecessor at the Bundesbank, Sabine Lautenschläger, already spoke at length about the imminent launch of the Single Supervisory Mechanism yesterday. I would therefore like to begin by just briefly outlining the key facts about the upcoming "sea change".

• In the second half of October the results of the comprehensive assessment will be published.
• The European Central Bank will assume responsibility for banking supervision in November. It will directly supervise the 120 most significant banks, including 21 German institutions. The ECB can also take on supervision of smaller banks. This is appropriate, as even these institutions can give rise to risks, for instance if they have similar business models and are hit by a common shock. The savings and loans crisis in the United States at the end of the 1980s and the Spanish cajas crisis are striking examples of this.
• The Single Supervisory Mechanism is just one pillar of the banking union, however. The Single Resolution Mechanism (SRM), which is scheduled to enter into force at the start of 2016, will introduce a European resolution authority and a bank resolution fund.

1. What are the aims of the banking union?

The emerging banking union comes with many expectations:

• the integration of the financial markets is to be promoted,
• risks of banks and governments untangled, and
• the stability of banks improved.

Similar expectations were attached to the introduction of the euro. The Delors Report of 1989 already highlighted that in a single market for financial services within a monetary union, a single monetary policy could make a positive contribution to financial market integration.²

I consider the banking union to be an important and necessary extension of monetary union. It closes a vulnerable gap in the single currency by harmonising rules for the supervision and resolution of credit institutions at the European level and establishing European supervisory authorities. This is important, regardless of whether you wish to continue monetary union in the spirit of a "Maastricht 2.0" with decentralised responsibility for fiscal policy or to develop it into a fiscal union with mutualisation of liability and control.

In the Eurosystem, the ECB provides banks with liquidity in its capacity as lender of last resort. Supervision organised purely along national lines is no longer fitting with the times. Liquidity assistance from national central banks can have an impact on the single monetary policy. The restructuring or resolution of financial institutions can overstretch the individual member states' financial means.

The banking union does not mean an extension of joint fiscal liability, however - its aim is anything but that. Rather, the banking union can play a role in establishing the liability principle at the European level in a better way than before. In future, private investors will have to shoulder losses. By contrast, public funding mechanisms may only be used in very tightly restricted exceptional cases, for example if the stability of the entire financial system is at risk. And if public funds have to be used, these should be national funds in the first instance.

In this sense, the comprehensive assessment serves to ensure that apparent legacy risks do not fall within the scope of the new supervision and resolution regime. They should instead remain where they arose - at the national level and under national control.

So is the banking union the solution to our problems?

It could be an important part of the solution. But further steps are needed to make it a success.

• First, we must - as previously addressed - ensure that private investors are adequately liable for losses incurred.
• Second, the banking union on its own is not the key to better separating the risks of banks and governments. We must also make further progress in regulation and put an end to the preferential treatment afforded to government debt instruments.
• Third, earlier reforms of the financial markets in Europe have focused on the banks. It should not be overlooked here that the integration of the markets for capital can make an important contribution to sharing opportunities and risks in monetary union.

I would now like to proceed by discussing how the banking union can contribute to better identifying and handling risks. But first I shall address the question of how opportunities and risks are distributed across integrated financial markets.

2. How are opportunities and risks distributed across integrated financial markets?

The European project is globally unique. Sovereign states have committed to integrating their markets for goods, labour and capital. The integration of the financial markets has made progress in many respects. In the years leading up to the outbreak of the financial crisis, European banks' claims within Europe had grown from 36 % of GDP at the end of the 1990s to 77 % in 2008. Since then, however, these claims have fallen back to around 48 % of GDP.

But the strong increase in banks' cross-border exposures masked the fact that the integration of the markets remained incomplete.³  In the banking sector, integration concentrated on the interbank market; credit markets for enterprises and deposit business remained predominantly national. In the big European countries, banks are mainly domestically owned.⁴  The integration of the markets for capital may have increased in Europe, but the ownership structures of many enterprises are nonetheless strongly national. Improved market integration is hindered by differences in national taxation and legal systems as well as by varying market practices. Political resistance to cross-border ownership structures is another factor.

The relatively low level of integration in the markets for capital makes it difficult for risks to be assumed via the financial markets. A comparison with the United States shows that equity holdings there are dispersed much more widely throughout the entire country. If a negative shock hits an industry or a specific region, then this loss is spread widely beyond that region. The same applies to positive developments, which are reflected in higher dividends. Equity investors participate directly in economic risk and in gains and losses. Creditors, on the other hand, are not exposed to losses - except in the case of insolvency.

Empirical studies for the United States show that integrated markets for capital cushion around 40 % of the cyclical fluctuations between the US federal states. A share of around 25 % is smoothed via the credit markets. Fiscal mechanisms cushion just 20 % of shocks.⁵ Studies for Canada and Sweden as well as the German federal states come to similar conclusions.

The integration of banks in Europe has certainly had positive effects on growth and investment. But the risks arising from integration were frequently overlooked. What would happen if assumed risks became loss events, if losses materialised, if European banks were hit by a bigger shock? None of these questions were seriously posed until the crisis struck.

During the crisis, it became clear that many banks had taken on excessive risk. That's why they ran into funding problems almost simultaneously and had to offload exposures. The freed-up capital often provided inadequate protection against the remaining risks. Up until the crisis, however, there was no special insolvency law for banks that allowed creditors to be bailed in and took account of risks to financial stability.

Fiscal and monetary policy had to step into the breach. The outcome is higher public fiscal debt and greater risks in central bank balance sheets. Some banks are still lumbered with legacy exposures and can only play an inadequate part in the necessary structural change in the real economy.

In this situation, the banking union can play a significant role in better identifying risks in the future by way of the Single Supervisory Mechanism. At the same time, the Single Resolution Mechanism represents a considerable step towards improving the manner in which risk is handled in a monetary union made up of sovereign states.

3. Can risks be better identified in future?

Identifying risks better is the aim of the Single Supervisory Mechanism. It was designed to roll out uniform supervisory standards across all participating member states. The Single Supervisory Mechanism will thus make an important contribution to better integration of the markets, which is currently being hampered, amongst other things, by inconsistent supervisory standards.

One of the major advantages of the Single Supervisory Mechanism is that it obtains comprehensive information. All supervisory reporting data for the most significant banks will thus be accumulated by a central authority in future. This information, which previously only existed at the national level, can be bundled to facilitate cross-border comparisons. In this way, the supervisor should be able to identify risks faster and better in future.

The Single Supervisory Mechanism also reduces the risk of supervisors developing an all too close relationship with "their" banks. Local knowledge is certainly of great importance in the banking business. But we should also keep an eye on possible misaligned incentives in national supervision. Experiences in the United States show that local and national supervisors apply uniform supervisory standards with varying levels of strictness.⁶ This is why we should do our utmost to apply rules uniformly in Europe, thereby creating a genuine level playing field for banks.

There are also areas in which the institutional framework can be improved for efficient European supervision.

• First, different tasks are bundled in the ECB. Alongside its actual task of monetary policy, the ECB Governing Council is ultimately also responsible for supervisory issues. These relate to microprudential and macroprudential supervision, with the latter addressing the stability of the financial system as a whole. If, however, monetary policy decisions are influenced by supervisory considerations, this would significantly impact on the credibility of monetary policy.
• Second, the non-euro-area countries in the ECB Governing Council are not entitled to vote. This hinders their inclusion in the banking union - and thus the integration of the banking markets in Europe as a whole. Central and eastern European countries, in which foreign banks have large market shares, are affected by this in particular. Clear-cut inclusion of these countries in the banking union would therefore be sensible for the creation of a single banking market.
• Third, the Single Supervisory Mechanism is extremely complex. A stringent organisational structure with clearly defined processes and reporting channels could alleviate the ensuing frictions, but certainly not eliminate them completely.

We ought to keep a watchful eye on these problems. Practice will show whether they can be satisfactorily resolved within the existing legal framework. If this is not the case, then we should make the necessary adjustments.

4. Will the banking union improve how risks that materialise are dealt with?

But having a uniform set of standards in banking supervision is not enough. We need procedures for dealing with banks that run into difficulties. What will happen in the future if the ECB, as the single supervisor, withdraws a bank's licence? How can we ensure that individual weak banks are able to exit the market?

The second pillar of the banking union, the Single Resolution Mechanism, was designed to provide the answer to these questions. National solutions can normally be found for smaller ailing banks. In Germany, for example, there were a good 4,700 banks in the market at the start of the 1990s - today there are around 1,800. The bulk of these changes came about through mergers of smaller institutions with economically stronger ones.

There are hardly any tried and tested mechanisms for restructuring and resolving major banks, however. Risks to financial stability can be posed if key business areas fail. Furthermore, coordination across national borders is required.

These problems are not easy to resolve. All too often, stressed banks were given too much time to resolve the economic problems they faced. In earlier crises, finance ministers and supervisors often acted too late and too indecisively. Some shied away from the consequences of banking crises for government budgets, others from the loss of reputation for supervisors. Empirical studies indicate that banking crises come at a high real economic and fiscal cost.⁷ A wait-and-see strategy is the most expensive approach of all, however.

That's why the Bank Recovery and Resolution Directive (BRRD) is an important step towards better handling troubled banks. It harmonises the legal framework for bank resolutions in Europe. The BRRD defines the liability cascade: shareholders are liable first, followed by the creditors of subordinated and unsecured bonds. Depositors are protected up to Euro100,000 by the statutory deposit guarantee scheme.

If the capital freed up in this way is insufficient, funds from the single European resolution fund can be used, under certain conditions. This fund is intended to hold Euro55 billion. It is to be built up over time and funded by bank levies.

The exact structure of this bank levy is currently being negotiated. Regardless of the details, it is particularly important in my view that the bank levy is as clear and simple to calculate as possible and does not allow for any exceptions.

To ensure that liability and control are kept in balance, banks that are subject to European supervision also need to be resolved at the European level. But this also means that losses originating from the time before the launch of the banking union need to be remedied under national responsibility.

With regard to resolution financing, national fiscal backstops and - as an ultima ratio - the European Stability Mechanism (ESM) are of central importance. In the case of legacy risks, it should also be taken into account that direct recapitalisation of banks by the ESM breaches the liability principle. After all, the legacy risks arose on the watch of national supervisors.

Practical application could reveal that subsequent improvements to the current institutional structures might be needed in two areas to ensure that the new resolution mechanisms are effective.

First, the decision-making structures of the SRM - much like those of the Single Supervisory Mechanism - are extremely complex. The short deadline of one weekend will hardly be sufficient for a restructuring or resolution to be initiated at a large bank in a way that does not place an excessive strain on the system as a whole. For this reason, a check should be carried out on the extent to which the legal framework needs to be altered in order to arrive at more efficient decision-making structures in the resolution mechanism.

The second problematic point is the authorities' relatively broad scope for discretion, which allows them to exclude private creditors from loss-sharing. This assessment is not an easy one to make.

• The higher the losses assumed by private creditors, the greater the potential negative effects for the stability of the financial system.
• The lower the private loss absorption, however, the higher the costs for government budgets - and the lower the disciplining effect for investors as well.

The resolution authority therefore faces a conflict of interests: if it aims to deflect risks to financial stability, it must trust that ultimately a fiscal backstop will be available.

The American model of systemic risk exception is of interest for implementing the liability principle and permitting as few exceptions from the bail-in of creditors as possible. The principle of bailing in creditors can only be deviated from in systemic crises. Each deviation must be approved by a clear majority of the relevant decision-making bodies - in this case, the deposit guarantee scheme and the central bank. The finance minister must sign off a deviation in consultation with the president. This is intended to secure the financing of bank restructurings. This is a sensible approach to strengthening the credibility of resolution regimes and being capable of acting during systemic crises at the same time.

One point is of particular importance to me here: fiscal backstops are not the path towards to greater mutual liability. Fiscal means can only ever be the last resort within a well-defined liability cascade. But without this last resort, there is the risk of bank recoveries or resolutions being delayed. This is why it must be ensured in advance that fiscal means are available to plug any capital shortfalls in an emergency. Only then will the resolution authority at the head of the decision-making chain take on the risk of resolving a systemically important institution. This will make creditor bail-ins the rule and not the exception. This, in turn, is the prerequisite for risks being adequately priced by lenders.

5. Where is further economic policy action needed?

I mentioned three aims of the banking union at the beginning of my speech: more strongly integrated and also more stable financial markets, as well as the need to disentangle the risks of banks and governments.

The banking union alone will not be able to achieve these aims, but it is a key building block that must be augmented and soundly anchored. I would like to emphasise three areas.

• First, credibly separating the risks of banks and governments requires further regulatory action plans. The banking union by itself does not change banking regulation. In future, the resilience of institutions will be strengthened by Basel III and the additional capital requirements for systemically important financial institutions. But this is not enough.

The preferential treatment afforded to government debt instruments needs to be put to an end in the medium term. Sovereign bonds need, like other bank exposures as well, to be backed by capital. The existing limits on large exposures should be gradually extended to cover sovereign debt.

• Second, deeper integration of the markets for capital in Europe should be made possible. After all, cross-border investment allows opportunities - and risks - to be better shared. This strengthens the resilience of the financial system. Despite the essentially free movement of capital, there are still legal and institutional barriers in Europe.
• Third, the banking union can accelerate integration of the financial markets by introducing a common set of supervisory standards. But above all, it can also contribute to better risk assumption by the private sector. The Single Resolution Mechanism in particular is expected to make private investors participate in risks that materialise. But for this to become reality, the new rules need to be rigorously applied, and exceptions to the bail-in of creditors must be minimised.

On the whole, the banking union means that the authority for supervising banks will be transferred to the European level. The need for this is nothing new - the Delors Report already pointed out exaggerations that financial markets are prone to, and the associated constraints that this entails for national economic policy.⁸

Thank you very much.

---