Andreas Dombret: Are we done now? Reflections on the post-crisis supervisory and regulatory regime

Speech by Dr Andreas Dombret, Member of the Executive Board of the Deutsche Bundesbank, at the University of Cape Town, Cape Town, 1 September 2016.

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

1. Introduction

Ladies and gentlemen

After roughly eight years of post-crisis reforms there is one simple question that needs to be asked: Are we done now? According to some newspaper articles and the opinions of a number of banking experts, the burden of strengthened regulation and supervision that has been imposed on today's Europe should serve as ample evidence. Admittedly, the post-crisis reforms were and will remain an arduous task for all concerned. The Basel III rules from 2010 alone tip the scales at a massive 616 pages. On top of this, the European Central Bank newly hired around 900 employees for the new supervisory regime set up for the euro area and around 4,700 supervisors from national authorities all over the euro area had to be coordinated to serve this new institutional setup^.1 Believe me - I would prefer a (much) lighter workload. But for supervisors, less work today often means a lot more work farther down the road.

But clearly, the costs of regulation and supervision alone are inappropriate indicators of over- or under-regulation and supervision. Even as seen from the banks' perspective, cries of overregulation are too simplistic since they neglect the potential long-term benefits of these reforms for banks, such as increased strength and trustworthiness. But, more importantly, what ultimately counts is how the costs and benefits of regulation and supervision affect society as a whole. Costs such as the huge redistribution from taxpayers to financial entities and individuals as a result of the financial crisis have to be part of any meaningful benchmark. With the benefit of hindsight, we may calculate costs and benefits. But unfortunately, this does not tell us how to identify optimal regulation and supervision for the future.

But is there possibly a better yardstick with which to identify optimal regulation and supervision? Solvency is a natural candidate. Safeguarding solvent institutions is a crucial element of regulation and of supervision in our current financial architecture. Sufficient capital makes it more likely that a bank will survive economic downturns, poor strategic decisions or idiosyncratic problems. In fact, finalisation of the current Basel III framework aims at safeguarding solvency of institutions by adjusting capital requirements. This involves large-scale quantification of risks for all aspects of banking, from bank lending and trading to operational perils faced by a bank. With respect to Basel III, the final agreements are expected to be ready by the end of this year. Generally, a lot can be learned from analysing data and looking at past experiences. But does this imply that we should eventually be able to answer the question "Have we done enough?" with certainty? If you ask me, this is a hazardous interpretation of the question. By this I do not wish to demean the value of calculations and data-based assessments such as those accompanying Basel III. It's more a matter of making sure they occupy the right place in our mind-set. Therefore, my speech will explore the progress we have made towards achieving our idea of a regulatory and supervisory universe.

To broaden the focus, I will start by presenting financial stability in an evolutionary framework. After that I will address the financial reforms introduced in Europe since 2008, looking at them through the lens of evolution in order to point out reforms that go beyond quantitative adjustments.

Let me express my pleasure at being invited to present these thoughts to the scientific community here at Cape Town University. Scientists are known for testing politicians and legislators against the goals they have set. But it is also essential for scientists to engage in the debate about methodology and overarching goals. If we do not conduct the discourse on financial stability in an impartial manner, it is soon likely to be governed by short-sighted and narrow-minded arguments again.

2. Beyond optimal regulation and supervision: approaching financial stability after the crisis

Optimality is an attractive analytical concept. But for good reasons, it never completely ruled policy for the financial sector.

First of all, the complex nature of financial markets, products and entities prevents us from identifying the optimum. When calibrating policies towards a given optimum, it is necessary to combine models and numbers. But numbers can be incorrect and models can be wrong. Complexity may even give rise to surprising side-effects. The issues stemming from our complex financial world are well-documented by now. Andrew Haldane gave that memorable speech about the dog and the Frisbee - it covers the topic almost exhaustively.

But the financial sector is not just complex. Dozens of crises over the past few decades were followed by quick policy responses, only to then be followed by some other sort of crisis. This experience has clearly documented that banks and other components of the financial sector are complex and adaptive systems - they are alive and kicking, so to speak. Just like living organisms, they respond to changes in the environment such as macroeconomic variables or new regulation by behaving differently. In economic terms, incentives respond to a change in circumstances. This in turn changes observable patterns in the financial sector. The search for yield episode immediately prior to the crisis is a typical illustration.

However, for the sake of completeness, undesirable incentives are not limited to bankers. Supervisory authorities themselves have been prone to bias. Some keywords here are regulatory capture, gold plating and supervisory negligence. During the euro crisis, quite a few instances of ineffective supervision came to light. Why did that happen? Evolution theory tells us: Because it was possible. One simple conclusion from the financial as well as the Euro crisis was that rules will be effectively enacted and followed only if they are "dynamically stable rules". Or, if I may translate this into bankers' English: People will abide by their good intentions or by the formal rules only if we can make it so that it is in their best interests to do so.

This already makes the political design of rules and institutions a difficult task. But let me stress a third issue with regard to optimal regulation and supervision. It concerns the undisputed aim of all political efforts in this sector: Financial stability. What exactly is financial stability? We can't put a finger on it. Instead, as it turns out, financial stability may be a somewhat elusive goal. It is only abstract in nature. As with personal health its relevance only becomes clear when we are in a bad state. There are many sources of financial instability including market turmoil, bank runs, or even hoaxes that lead to (public) distrust in banks. Accordingly, we can make a distinction between policies that decrease the probability of a bank failure and policies designed to make an individual bank failure as bearable as possible and there again policies geared to addressing sources of distrust. Alongside such measures there is a need to allow competitive forces to prevail wherever appropriate in order to foster profitable and therefore resilient banking as well as to potentially stabilise economic conditions. All of these aims are connected, but they are not the same. Hence, financial stability policies may target different aspects of financial stability.

As a consequence of the arguments mentioned above, it is futile to lay claim to an optimal solution for regulatory and supervisory activities, because these deal with a constantly evolving system along with an abundance of complex, partly unforeseeable interactions and, possibly, loopholes.

The lessons learned from past crises might at first appear quite restrictive but they in fact remind us not to cling to the illusion that we have all the answers and absolute power. Instead, we are always bound to be one step behind and may even sometimes be wrong-footed. That said, there is some good news for researchers on financial regulation and supervision: Your job is safe as there will always be a constant flow of new developments in the financial sector for you to research.

So how do we move forward? Evolutionary theory not only highlights the problem, it also illuminates possible strategies. One of these concerns the role of knowledge in an inherently uncertain world. For sure, some models of the financial world and logical explanations for past regulation have failed. Such failures have arisen from model risk, in other words the fact that reality is (often) different from assumed causalities. And yes, there is also a risk of financial players taking unfair advantage of our inclination toward modelling financial risks. At the same time we must be careful not to fully reject theories and extrapolations about finance. Model-based hypotheses and data-based research are still valuable and they are certainly better than an erratic change to regulation or supervision. For example, I remain convinced of the basic concept of "value at risk" when calculating capital requirements. What we have to admit is that these numbers can still be falsified - and this most certainly will happen at some point in the future. Ultimately, quantitative adjustments such as those witnessed with respect to Basel III form part of a bottom-up evolution of the financial stability regime.

Beyond quantified requirements, there is little scope for definite answers. Such certainty would require a concrete idea of picture-perfect finance. Instead, we need to employ the whole arsenal of procedural wisdom including the application of best practices, multiple lines of defence, institutional rivalry, hierarchies of goals, adaptable rules and principles. This basically means that in order to achieve the elusive goal of financial stability, we have to look out for effectual processes rather than perfectionist end states.

3. Supervision and regulation in the EU - merely adapted to the latest crisis or fit for the future as well?

How are we to interpret our regulatory and supervisory regime against this background? Development in Basel in the recent past should be viewed as part of a bottom-up evolution of the financial stability regime. In spite of the inclusion of a huge mass of statistical data in the calibration of capital requirements, the final results will remain hypothetical and a political compromise. Safeguarding solvency and liquidity is just one of a number of strategies aimed at creating a robust financial framework. Another core element of post-crisis regulation was to make rules coherent, internationally consistent and to avoid bypass. Let us not forget that financial actors and entities, as living organisms, will eventually discover and make use of any available loophole. If there is any way to evade short-term pain, an industry worth trillions of euros will always look for it. When I say this I am not expressing criticism, I am simply stating the facts. Bypassing the rules is a key reason why regulation has become even more complex.

But there have been more lessons learned. One important realisation has been that a robust banking sector requires more than one line of regulatory and supervisory defence. It is true that solvent banks are central to financial regulation and supervision. However, in place of one single "perfect" line of defence, where success depends one hundred percent on these legal ideas and supervisory processes, we now rely more heavily on multiple lines of defence. Taking solvency and liquidity as a starting point, supervisors have intensified their efforts to target the root causes of the problems. Capital buffers were introduced which increase banks' room for manoeuvre in a downturn. Stress tests examine how robust a bank's capital is in terms of adverse economic projections in the future. What is more, eurozone supervisors are paying greater attention to business models as today's strategy feeds future risk capacity. Supervisors have even sharpened their focus on the culture and behaviour of bank officials. There is in fact a link between the behavioural inclinations of these officials and actual solvency: All of the steps taken are geared to minimising the likelihood of deterioration and failure at the earliest possible stage.

But solvency is not the only pillar of a stable financial sector. The European resolution framework has complemented solvency requirements. For even in cases where all previous efforts to keep a bank alive and healthy prove unsuccessful, the financial system should not be prone to panic reactions and contagion effects. Banks and responsible authorities have to be prepared for failure scenarios and guarantee smooth resolution. Importantly, resolution and solvency policies are not aimed at identical issues. In fact, crisis management maintains financial stability in a different way, namely by making sure that bank difficulties do not negatively impact private savings or market discipline, something which cannot be achieved effectively through solvency regulation.

There are several ways to read the most extensive institutional reform in Europe recently, the so-called banking union. Clearly, the euro zone faced serious distress, and our institutional framework had not proven credible nor effective - market reactions showed this quite unmistakably. Through the lenses of evolutionary adjustments, the move towards the banking union was a patch to our historical supervisory system, because it, too, had shown to be a living organism that had reacted in an undesireable way to the financial crisis. The most developed part of this banking union is the Single Supervisory Mechanism, which has rearranged the supervision of 129 of the largest and most significant institutions in the euro area, thus improving supervision in a number of ways. Given the fact that even supervisors are prone to error or even bias, the mechanism builds upon international teams under the auspices of the ECB, with the favourable effect that national interests are dampened. On top of this, supervisors' hands are tied to a greater degree as the setup makes it easier to compare supervisory measures and results. Last but not least, all supervisory and organisational decisions have to be approved by a common board, so that all national competent authorities have their say in the joint architecture. In everyday-life, the vast amount of decisions consume quite a lot of resources in national authorities. But from a financial stability point of view those institutional frictions and even opposed interests are manageable inasmuch as institutional rivalry may enhance checks and balances. In sum, the Single Supervisory Mechanism has changed supervisory incentives. It remains to be seen whether another pillar of the banking union, the Single Resolution Mechanism, will improve effective resolution in a similar fashion.

4. Conclusion

Ladies and gentlemen,

I began this speech by posing the question: Will the post-crisis regulatory and supervisory regime a project be finished soon? My response to this question is that while the Basel III reforms are set to be finalised by the end of the year and while there are no further reform plans in the policy pipeline, I feel we should frame the overall question a little differently. In my view, the supervisory and regulatory changes witnessed in Europe and in many other parts of the planet do not represent a response to a call for "optimal" stability, but they have set the basis for an evolutionarily fit regime. It is not just a matter of quantity, as suggested by current arguments about calibration and by newspaper articles focusing on over-regulation.

Confronted with an adaptable financial system, important improvements in supervisory and regulatory regime include:

• addressing incoherent regulation across countries and closing regulatory loopholes
• extending our supervisory lines of defence to circumvent the risk of putting all our eggs in one basket
• exploiting the possible synergies of the supervision in the Euro area to make sure that existing incentives match the original intentions

The evolutionary approach to financial stability amounts to more than an academic dry run. It shines a light on regulatory strategies which have come about as a high-priced lesson from past crises. Financial stability is not a narrow concept and it cannot be achieved through a singular, ideal set of policies and institutions. There is constant evolution wherever we look. Instead of directing our energies to identifying some unknown optimal state, we need to strive for plausible processes and a robust set of institutions.

The current narratives criticising over-regulation and over-supervision need to reflect these aspects as well. For representatives of the financial industry who are suspicious of overprotective administrative bodies as the force behind recent reforms, the evolutionary perspective should offer some alternative explanations.

Suffice to say, the precautionary measures that have been taken do not render criticism from the banking industry entirely meaningless. Input from the industry is welcome, as this helps to better understand the permanently evolving financial system. But we should be wary of distorted arguments based on partial calculations and a selective approach to the various pros and cons. Rather than looking at the status quo and asking whether there has been too-much or too little regulation and supervision thus far, we should recognise that we have a common interest in achieving better regulation. This is not a matter of mere quantity, but of quality and prudence, too. Just as our own health is not defined solely by the number of pills we take, the health of the financial sector, alive as it is, cannot be safeguarded by simply prescribing a narrow set of financial and supervisory remedies.

Thank you.

---