Andreas Dombret: What is "good" regulation?

Speech by Dr Andreas Dombret, Member of the Executive Board of the Deutsche Bundesbank, at the 16th Euro Finance Week, Frankfurt am Main, 19 November 2013.

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

Central bank speech  | 
25 November 2013

1. What is "good regulation"?

I would like to thank the hosts of the Euro Finance Week for raising this important topic of "good regulation" - even though fifteen minutes is not very much time to give a comprehensive report on the topic. The first question that occurs to me in this context is: who is regulation supposed to benefit in the first place? Regulated parties are likely to have a different take on the matter to supervisors and society at large. Instead of focusing on the theory, I therefore want to take a pragmatic approach to the topic, and to explain three key aspects from the regulators' point of view - although I cannot, of course, give an exhaustive account in this speech.

First, "good regulation" of financial markets pursues a clear objective, in my opinion: to create a sound framework for the financial system and thus ensure its stability. A stable financial system ought to be able at all times to fulfil its core task of allocating capital efficiently. In other words: efficient funding of the real economy should be constantly assured. That is the main yardstick we should use to measure regulation.

To keep the financial system stable, it is therefore important to apply a fundamental principle of market economies - incentive and punishment mechanisms - to regulation: risk-taking and liability have to be consistent with one another. This means that those who take risks in order to make profits also have to shoulder the risks involved - irrespective of their institutional form or size.

2. Put an end to "too-big-to-fail"

This principle, however, was nullified during the crisis. Several banks facing the threat of bankruptcy had to be bailed out by governments with taxpayer funds. These banks were regarded as too complex, too interconnected or simply "too big" to fail. There were fears that their failure would have a dramatic effect on the financial system and the real economy. Regulatory policy considerations took a back seat, and taxpayers were ultimately on the hook for risks on banks' balance sheets. The implied government guarantees for such institutions, which were already thought to exist before the financial crisis, were thus de facto confirmed.

That goes against the principles of a market economy. Private funds should be used for the recovery of a distressed bank. If shareholders are not willing to take that step, the institution should be resolved - without destabilising the entire financial system in the process. The option of a credible resolution is the only way to eliminate implied government guarantees for systemically important banks.

We are on the right track with regard to dealing with the systemic importance of banks, but the finish line is still a long way off. The new international agreements on resolution regimes developed by the Financial Stability Board (FSB) are a crucial step. Applying them globally will make the orderly resolution of a big bank a more realistic option and thus also a more credible threat. Intensive work is being carried out at the European level on the implementation of these principles. Negotiations on the European Recovery and Resolution Directive are scheduled for completion by the end of this year.

In Germany, the Bundestag already adopted a restructuring act in 2010 which anticipated the European rules. That was followed up by German lawmakers with the passage of the Act to Strengthen German Financial Supervision in 2012. Not only the new resolution regime but also, for instance, higher capital requirements, or the obligation to trade derivatives via central counterparties, represent a distinct improvement on the situation before the failure of Lehman Brothers. These are all key first steps, but they are not enough to make the financial system a safer place.

However, it is not only banks that can be systemically important. In July 2013, the Financial Stability Board published a list of global systemically important insurers, or G-SIIs. Much like systemically important banks, these nine institutions will be made subject to recovery and resolution planning requirements and capital surcharges. These measures, too, are designed to strengthen market economy principles.

It is clear that we have to shape and transpose the new rules agreed at the international level consistently across national borders and sectors alike. This is not such a simple matter in practice. However, it is the only way to create a "level playing field" and prevent regulatory arbitrage. And that is also the only way we can make progress in containing the "too-big-to-fail" risks.

3. The banking union - an important and welcome step

In my opinion, strong institutions are the second component of good regulation. Good regulation requires more than just well-thought-out legislation; good institutions are crucial. And this brings me to the European banking union. The single supervisory mechanism (SSM) has finally been adopted, and the European Central Bank (ECB) has begun to set up the necessary structures - including recruiting the staff required. The Bundesbank is supporting the ECB in these efforts.

The aim of joint supervision is for banks everywhere to be supervised according to the same high standards. In addition, cross-border effects can be covered better through joint supervision than by national supervisors. The financial system is likely to become more stable under European supervision. But, if push comes to shove, banks supervised at the European level will have to be resolved at the European level, too.

For this, we need a centralised European resolution authority for banks which is based on sound legal foundations. The debate on this single resolution mechanism (SRM) is in full swing, and several proposals are on the table. I believe the best solution is one that promises to be effective and enforceable in a crisis event while at the same time avoiding potential conflicts of interest on the part of the resolution authority from the outset. It is clear to me, too, that the SRM for banks should be launched as soon as possible after the SSM. I am for the beginning of 2015 as the kick-off date.

4. Strengthen macroprudential supervision

The third and final aspect of "good regulation" I wish to mention is the macroprudential perspective. After all, "good regulation" must never lose sight of the "big picture". The financial crisis swept away any lingering doubts that we need to broaden our regulatory perspective beyond individual banks. This means that we need macroprudential oversight alongside microprudential supervision.

This has now been enshrined in German law: the Financial Stability Act has been in force since the beginning of this year. The institutional framework for macroprudential oversight has thus been established in Germany. The newly formed Financial Stability Committee started its work at the beginning of the year. The Bundesbank is a member of this committee and has been tasked with the ongoing analysis of developments relevant to financial stability and with identifying threats. Where appropriate, we can propose that the committee issue warnings and, if necessary, specific recommendations for action.

5. The "cobra effect" and the preferential regulatory treatment of sovereign debt

Having mentioned these, to my mind, important three aspects of "good regulation", I would like to touch briefly on a core problem: complexity. In my opinion, complexity is one of the central challenges in both the design and the implementation of new - and, hopefully, good - regulation. And nobody would dispute the fact that financial markets and their participants have become very complex.

There are many levers which supervisors and/or lawmakers can pull. In some cases, pulling only one single lever the wrong way is enough to threaten the stability of the system. This can lead to unintended consequences such as regulatory gaps, duplication or arbitrage.

Let me discuss, at this juncture, a phenomenon known as the "cobra effect". The cobra anecdote goes back to colonial India. In order to contain a plague of cobras, the British colonial administrators offered a bounty for every dead cobra. The plan initially seemed to work; the colonial administrators were presented with a large number of dead cobras. Unfortunately, however, this failed to bring the plague under control. Most of the snakes delivered were actually not wild animals, but had instead been bred by entrepreneurial Indians in order to collect the bounty. Once the governor caught wind of it, he cancelled the premium, and the breeders then released the now-worthless snakes into the wild - thus making the plague worse than before...

Regulatory measures, too, can have undesired and unintended consequences. When the first Basel Capital Accord was introduced 25 years ago, industrial countries´ government bonds, for instance, were classified as safe because nobody thought a government could default. Bank regulators thus regarded government bonds as an anchor of stability for banks' balance sheets. As a result, banks today do not need to hold capital against the risks on OECD countries' government bonds. In addition, rules to limit concentration risk do not need to be applied to government bonds.

It is therefore no big surprise that this exception in the Basel Framework had unintended consequences. Banks poured massive funds into government bonds - especially domestic bonds. This created a kind of "doom loop" between the solvency of sovereigns and the solvency of domestic banks. We have to break this dangerous loop by ending the preferential regulatory treatment of government bonds.

However, as the story of the cobra effect shows, abrupt changes carry dangers too. Immediately prescribing that a certain amount of capital be held to back the risks on government bonds could lead to another set of unintended consequences. But that does not alter the fact that the existing rules need to be changed in order to break the "doom loop" between sovereigns and banks. Such change will never be popular, and it is clear that we need an appropriate transitional phase, such as by gradually phasing in new rules. Additional measures may be needed in order to avoid adverse effects on banks and sovereigns alike.

6. Conclusion

Let me conclude by summing up what I believe to be the indispensable elements of "good regulation". The aim of "good regulation" is to maintain the stability of the financial system, thus ensuring that the financial sector has the tools it needs to completely fulfil its role for the real economy. It is therefore important that market economy principles - such as the unity of liability and control - also apply to the financial sector. Moreover, strong institutions are needed to implement and apply regulation. Lastly, a macro perspective is needed. This means that we need not only the existing microprudential regulation but also a new, systemically-oriented regulatory framework. The two must go hand-in-hand.

On top of all this, we need to be highly vigilant towards potential design flaws and unintended consequences in all of these regulatory projects. And we also need the wisdom to make corrections if need be. After all, the financial system is evolving relentlessly. "Good regulation" will thus always be a work in progress.

This marks the end of my introductory remarks. I look forward to our discussion.