Ingves: The need for a strengthened insolvency and resolution framework for banks – A central banker’s perspective

Speech at the International Bar Association’s 14th Annual Global Insolvency and Restructuring Conference, 19 May 2008

As a central banker, the tasks assigned to me are to maintain price stability and to promote a sound and efficient payment system. In short, I deal with economics and finance. So what does this have to do with your business, insolvency law? What is a central banker doing opening this conference on insolvency and restructuring?

Actually, there are more links between the fields than you might think. As I said, one of my tasks is to oversee the payment system. Without a functioning payment system, no modern economy is able to function for very long. But what we call the payment system is really not much more than a system where money is transferred electronically from one bank account to another, in a longer or shorter chain of intermediaries with the central bank as a hub. If there is a problem in one of those intermediaries, there is a problem in the payment system. And if the problem is not technical but financial in nature, this may lead you right into the area of insolvency law and bank restructuring regimes.

But in fact, there is also a link of a more general nature. Maintaining price stability, which is the central bank’s primary objective, means that overall prices should be stable. That task is made more difficult if some markets are not working properly, for instance if there are “bubbles”, or misalignments, in asset prices. I think that we can never fully protect ourselves against this happening. But a first step is to ensure that the market has an incentive to assess risks properly. This can only be achieved if it is clear to the agents that they are always responsible for their investment decisions. To ensure this, you need well designed and credible insolvency frameworks.

So, what is the problem? Well, the situation varies between countries of course. But in many cases, quite frankly, I do not think that our present regimes for dealing with weak banks are adequate. In particular, I doubt that they will be of much use in a situation where the payment system is at risk. And moreover, I fear that our attempts to deal with the situation in the heat of the moment may require us to take action that will solve the immediate problem, but which may in the end reduce the long term market discipline.

What I call for is a special insolvency and resolution framework for banks and other deposit taking institutions. A system that works not only when there are minor banking problems, but also when the payment system is at risk. Such a regime does not presently exist in Sweden – in fact, many countries lack such a regime. Indeed, this gap in the regulatory framework has been demonstrated in the recent period of financial turbulence, where a number of countries have found it necessary to take extraordinary measures with regard to banks and other financial institutions in order to maintain financial stability.

Today, I will try to sketch the broad features of such a resolution framework. In brief, it is a question of finding a system that makes it possible to close banks quickly and efficiently at a low cost to the deposit guarantee system, while making shareholders and uninsured creditors bear the responsibility for their investment decisions. But it must also allow for special measures to be taken in the exceptional situation where the failure of a bank risks destabilising the financial system as a whole. Also in such situations the government needs to have a strong negotiating position against the shareholders.

But, first, how did I arrive at this conclusion? Well, from an analysis of the role of the banks in the economy and the safety net which has been built up around the sector. And from some experience from dealing with real crises. To get the right perspective, I think it is useful to start at this end.

What is so special about bank failures?
Bank failures have always been a problematic issue for governments. A bank failure affects not only shareholders and professional investors, but also all the ordinary people who have put their money in the bank in the belief that it will be safe there. It is not without reason that you often hear the reference to widows and orphans running the risk of losing their money in connection with bank problems.

And even if the depositors may in the end recover much of their money in a subsequent insolvency procedure, they may lose access to the money in their bank account, and the possibility to make payments from it, for quite some time. As most of us have most of our money in a bank account, we may find it difficult to quickly find the means needed to pay our rent and food if this happens. And you may also be affected by the failure of other peoples’ banks. For instance, if your employer’s bank gets into problems, the pay-roll account might be frozen and you might not get your salary. And if your employer cannot quickly get credit from another bank, he may not be able to pay his subcontractors and his utility bills. He may also find it difficult to manage the risks in his business, for instance in relation to trades in a foreign currency. At worst, he may need to shut down entirely. A bank failure can thus have negative effects on the real economy.
The close and complex financial links between the banks also mean that a failure can have a knock on effect on the other banks in the system. Particularly problematic in this respect will be the failure of a bank which has a key position in the payment system. That is, a bank which many other banks use as their intermediary when making payments to one another. The mere rumour that such a bank is having problems can thus paralyse the payment system.

A further complication is that banks are vulnerable in a special way. If confidence is shaken, deposits and other financing can usually be withdrawn at short notice. The assets can not be disposed of that easily. A bank may be forced to sell its assets at fire sale prices to meet a run; if the worst comes to the worst the bank may become insolvent. Banks are therefore very sensitive to a shift in confidence.

A bank failure can thus affect the smooth functioning of the financial system. The more widespread the problem and the larger the risk of the crisis spreading through the system, the greater is the threat to financial stability – and in the end to the real economy. Clearly, the failure of a bank where a lot of people, companies and other banks have their accounts, and which is an important provider of credit and risk management tools, will normally be more problematic than the failure of a small bank. But also the failure of a small bank will have wider effects, and even quite severe ones if it happens at a bad time.

Given these wider effects of any bank failure, a number of special regulations to deal with the banks have been introduced over the years. As a result, the banking sector is now among the most regulated industries in our economies. The safety net includes financial supervision to monitor the risk taking of banks, with regulations on how much capital a bank needs to hold and how it should manage its risks. The banks’ owners enter into the business knowing that a serious breach of these regulations should lead to a revocation of the licence and the bank being closed.

The safety net also includes the ability of the central bank to lend on special conditions to an illiquid but solvent bank when financial stability is at risk. In many countries, there is also an explicit guarantee to the small depositors, for which the banking sector itself is to take the responsibility.

But this safety net, while reducing the risks involved, provides no full protection against failures. In particular, it does not extend to the case when the financial system is threatened in its entirety. And while the formal depositor protection should reduce the pressure from the public on the government to intervene in a bank failure, if the scheme is not properly set up, this expectation may prove unfounded.

The government has a direct interest in an efficient insolvency and resolution framework
So the stakes are still high for the government in a banking crisis. While owners have little to lose when capital is low, and many creditors are protected by the deposit insurance, the state – and the whole economy in the end - may at worse suffer enormous losses. Therefore, unlike the situation in ordinary companies, the lack of an efficient insolvency regime will not primarily be a creditor problem only, but a problem which will fall right into the government’s lap.

And if the insolvency framework is not efficient and cost effective, it may be rational for the government to intervene beyond the ordinary regulatory framework. If there is no other way to deal with the problem, it may even be rational to contribute capital directly. This may be the case even if such an investment cannot be justified on the basis of a corporate capital investment appraisal. The alternative of not intervening may simply be very much more expensive.

Now, if this intervention by the government makes the owners and creditors better off than they would have been otherwise, the rules of the game change completely. If owners and creditors can speculate on a bail-out, market discipline will be eroded. Clearly, this needs to be avoided. And the conclusion is that, if anything, it is even more necessary to have effective and efficient insolvency and resolution frameworks for banks than for companies in general.

In the Swedish banking crisis, the public interest was secured by special crisis legislation
So, how do we fare in this respect? Well, let us take a look at some real crises.

I will start with the case with which I am most familiar, the Swedish financial crisis in the early 1990s. I got involved in the management of the crisis in my capacity as an official at the Ministry of Finance. At the time, the country had severe macroeconomic and financial problems, with a number of banks reporting losses. Moreover, the banks were finding it increasingly difficult to obtain financing in the international capital markets, a source of funding they were heavily dependent on. The international counterparties reduced their exposures not only to those banks that were having fundamental problems, but to Swedish banks in general.

In order to avoid a collapse of the banking system, it was necessary to take action. We tried several measures to contain the “silent run” on the banks. What proved effective was the declaration by the government that the banks would meet their debt obligations as they fell due. This general guarantee, as it has been called, did not take the form of a legally binding guarantee, but was merely a political commitment that the creditors would not suffer losses.

At the same time it was necessary to deal with the banks in distress. And it was quite obvious that we could not do this within the normal court-based insolvency procedure. This procedure would simply be too slow, and would not take into account the need to keep the operations running in order to maintain the functioning of the financial system.

With creditors insulated against losses by the guarantee, and with the formal insolvency regime assessed to be inappropriate, we needed to find another way to ensure that losses for which the shareholders were responsible would not fall on the tax-payers. In order to secure this, we needed to take control of the banks concerned before capital was completely eroded. Needless to say, we found ourselves in a rather poor bargaining position against the shareholders.

Fortunately, a Bank Support Act, laying out the principles for the possible actions, could be implemented quite swiftly. A special crisis management authority was also set up. Inspired by the US model, where the authorities can take control of a bank when capital falls below a certain level, this temporary crisis legislation gave the government the right to take over the shares in undercapitalised banks. The Bank Support Act also made it possible to tie conditions to the actual use of the guarantee. Thus, the main rule was that any financial support given directly to a bank would result in a proportional dilution of the original shareholders’ holdings. The Bank Support Act also laid the basis for arranging private take-overs where possible. And, not least, for the orderly winding up of unviable banks.

The measures imposed, while quite restrictive to the shareholders, were deemed as appropriate by the legislator and the judicial system in light of the wider public interest of maintaining financial stability at a low cost to the government and the tax-payers.

All in all, I think that it is fair to say that the outcome was not too bad, given the circumstances. I am also not too worried that the measures imposed have led to a weakening of market discipline in this country.

Nevertheless, this does not mean that we could not have done better if the appropriate legislation had been in place already at the outset of the crisis. Moreover, I am afraid that it may be more difficult to handle a crisis by drafting temporary legislation in the midst of a crisis today. First, the banking sector is more internationalised. In the Swedish case, many banks are part of large cross-border banking groups with strong intra-group links, which may not be easy to dissolve quickly. This means that it is not obvious that a crisis could be handled on a national basis. Second, Sweden is now a member of the EU. Any measure taken will thus have to be in compliance with EU legislation.

Sweden has actually also had recent experience of the failure of a deposit taking institution. The failure in 2006 of the credit institution Custodia again displayed the problems in relying on the ordinary corporate bankruptcy regime. Custodia was tiny, so there was clearly no systemic risk. But it highlighted a number of deficiencies in the legal framework. It turned out to be a long process to finally have the institution closed, something which may have contributed to the failure coming at higher costs to the deposit insurance scheme than necessary.

Other countries appear to have had similar experiences
Taking a broader outlook, it is not uncommon that countries rely on some modification of the ordinary corporate insolvency procedure to deal with weak banks, and in some cases a different regime altogether. And many countries seem to be able to handle small problems within their ordinary framework.

But when it comes to those exceptional circumstances where the payment system and the stability of the financial system are at risk, the common denominator between different crises seems to be that the authorities did not find it possible to rely on the normal rule. Instead there have been special provisions or special crisis legislation.

A case which I followed closely was the Asian crisis in the late 1990s. Like in Sweden in the early 1990s, it was necessary to take specific measures to avoid a serious interruption of the financial system. And there are a number of emerging markets which have had similar experiences and have been confronted with the same dilemma of how to secure financial stability without overly eroding market discipline in the long run.

But I think that until recently, most industrialised countries may have felt quite removed from this problem. The recent experiences in the UK, the US and Germany with extraordinary measures needed to deal with the crises in the building society Northern Rock, the investment firm Bear Sterns and the bank IKB respectively, is a reminder that these situations can, and in many cases have, surfaced anywhere.

They also seem to confirm that the measures that you wish to take may be in conflict with laws and regulations which were not drafted with the specificity of a banking crisis in mind, but nevertheless apply to banks in these situations. A telling example is the Northern Rock experience, where it was assessed that the preferred option to fix a private takeover might be difficult to arrange given the design of the regulations on insider trading and take-overs. More generally, there is also a debate on whether measures that restrict the rights of the existing shareholders might be in conflict with company law and even with human rights more generally, and on whether guarantees extended might be in conflict with state aid rules.

I want to make it clear that I have no views on market abuse regulations, regulations on shareholder rights in company law or state aid rules as such. They fulfil an important function and as far as I can see usually work well together. The fact that human rights must be respected can hardly be called into question. And I wish to emphasise that I do not call into question the role of the EU in the regulation of our markets. On the contrary, it is important to deal with these issues on an international level.

But the overall framework has not been drafted with the specific situation in the banking sector in mind. It may therefore be difficult to find suitable solutions in a crisis situation without running the risk of breaking a number of regulations. It is not acceptable that decision-makers should have to fumble in an acute crisis situation, trying to steer clear of this risk. It is also important that the action by the authorities in such a situation is evaluated against a regulatory framework fit for its purpose. A framework which makes it possible to maintain a proper balance between the different parties involved also in a crisis situation.

There is a need for a special insolvency and resolution framework for banks
In my opinion, all this points in one direction. A special insolvency regime is needed for banks and other institutions that are covered by the safety net.

In this country it has long been said that banks should be covered by the ordinary corporate bankruptcy legislation, at least as long as there is no direct threat to the system. This is because government intervention over and above the existing safety net for the banks should be avoided.

I believe that this is the wrong way to look at it. The ordinary bankruptcy regime does not sufficiently recognise the need for speed when dealing with banking problems. Moreover, the techniques normally used, such as a stay on payments, are ill adapted to banks. To supplement financial supervision and the deposit guarantee with a special insolvency regime, would not be adding an additional safety net to the benefit of depositors. It is quite simply a way of ensuring that the safety net protects those it was intended to protect but no others, and at a reasonable cost to the deposit guarantee system and society as a whole.

But it should also be recognised that in situations where there is a threat to the financial system in its entirety, it may be needed to take partly different measures than in an ordinary bank failure. In these situations, the key focus must be on maintaining the vital functions in the interest of society as a whole.

Perhaps the idea to have a partly different framework in these situations seems odd to you? After all, we normally expect the same legislative acts to apply regardless of the state of the economy or the situation in the financial system.

But the situation is not unique. A similar situation exists in the health sector, where society has set up a framework to deal with our usual flus, broken bones and heart attacks. There are doctors and hospitals to deal with our problems. And there may be guidelines on how long sick leaves will be needed for a certain diagnosis. But I think that it is also clear that in some severe situations, a different framework will need to be applied. In case of an epidemic disease, such as the avian flu, there may be a need to direct those infected to separate hospitals. And it may be necessary to establish an order of priority for the allocation of the scarce medical resources. One where the guiding principle may be the survival of the society as a whole. It may also be the case that the crisis regulations clash with the normal regulations. For instance, you are normally not entitled to take a sick day if you are actually healthy. Now, even healthy people may be asked to leave their offices to avoid the disease from spreading, while the key employees may be asked to go to work to keep the operations running. When the principles clash, it will be vital for society that the ones needed to contain the epidemic will prevail. And if that special framework is not already in place, the politicians will need to get involved.

Clearly, a bank failure cannot be compared to a serious injury or sickness in terms of personal suffering. But it can be an event which will make many people’s lives very difficult. And a systemic banking crisis, like an epidemic, may stall the functioning of our entire economies. It is therefore important to be able to deal with an ordinary bank failure as well as a more serious systemic crisis. And just like in the health sector, when the regulations clash, it must be clear which ones will dominate.

There is another similarity, I believe. Like with the flu, it may be difficult at first to tell the nature of the disease. It may not be obvious at first if it will be possible to deal with the problem bank within the normal framework, or if there will be a need to take special measures. Therefore, all cases should be dealt with within the same overall structure.

A speedy insolvency procedure with a fast pay-out to depositors at a low cost to the deposit guarantee scheme…
Setting up a framework to deal with an ordinary bank failure does not need to be overly complicated. Experiences from, for instance, the United States and Canada show that it is quite feasible to close banks that do not follow the rules without this leading to consumers losing confidence in the financial system, and at relatively low cost. The solution is based on the deposit guarantee authority having a right, in fact an obligation, to intervene in a bank when the capital falls below a certain level. One first ensures that the depositors covered by the guarantee have access to their funds. Then the authorities can choose from a broad arsenal of measures, including the transfer of the assets and liabilities to another bank or the merger with another bank, to solve the problem at the least cost to the deposit guarantee scheme.

I think that there are two characteristics of these kinds of systems which merit particular attention. First, the ability to take swift measures and to write down claims quickly. It is important to restore access to the liquidity held in the bank accounts as quickly as possible not only for the insured depositors, but also more generally. Second, the system is constructed such that the principle that owners take the first loss, and after this the creditors in any order of priority that might exist, cannot be circumvented. The swift intervention means that there is no point in running to the bank to “get first in line”. The fact that the possibility of a profit reflects a risk of a loss is fundamental in a market economy, and the system thus ensures that this principle is maintained.

…with robust financing of the deposit guarantee
A related issue refers to the need to set up a deposit guarantee scheme in such a way that the responsibility for the small depositors is fully internalised by the banking sector not only on paper but also in practice. In addition to the need for a quick pay-out to depositors to avoid the risk of a run, it is also important that the financing of the scheme is robust.

Many schemes rely on a mutual responsibility, where the remaining banks will come up with the financing to cover the insured depositors of a failed bank, in some cases relying on a fund which has been built up in advance. This works fine as long as it is only small banks that suffer problems. However, if a larger bank is affected, there is a risk that the problems will spread as other banks are forced to contribute substantial funds from their own balance sheets. Given that many markets are very concentrated, with a few banks dominating the market, the failure of the largest banks will be difficult to finance in these kinds of mutual systems.

Particularly vulnerable are the kind of small private schemes which rely solely on the remaining banks coming up with money after a failure. The deposit guarantee can then rather become a destabilising factor without the problem bank necessarily needing to be a particularly important actor in the payment system or on the financial markets.

Clearly, you want to avoid a situation where government intervention may be necessary simply because of a poorly designed deposit guarantee system.

A solution to the problem was put forward in a Swedish government enquiry a few years ago. The trick is to make the banks internalise the cost of the guarantee not at the industry level, but at the level of the individual bank. The way to do it is to set up a state-administered system to which all the banks pay a fee in relation to their risks, much like an ordinary insurance policy.
Sure, the state still stands the risk in the event of a large failure. But in this system, it is duly compensated for the risk. And the financing is robust. If implemented widely, this solution will also make for a more level playing field between banks from different countries competing partly on the same markets. I therefore hope that the proposal will get more international attention than it has so far.

There must be a strong negotiating position for the government even when financial stability is threatened
The above set up is a good set up to be able to handle the average bank failure. But, again, in some situations a different regime may be needed. In particular, this may be the case when it comes to larger and more complex banks in a situation of markets turmoil, when there is a systemic threat to the financial system. In this situation, it may not be possible to immediately revoke the licence for a bank which is in serious violation of the supervisory regulations. The focus will need to be on maintaining the key operations in the financial system. To secure this, the bank may need to stay open, at least for a while. The challenge is to secure also in these cases that the state does not end up with the losses that the shareholders and creditors should bear.

The first principle is to ensure that the bank’s owners will bear the first loss in a crisis. This does not need to be very complicated in practical terms. In principle it is a case of the supervisory authority having the possibility to quickly and efficiently write down the capital against losses and ensure that the owners’ control over the bank is reduced to a corresponding degree. Such a write-down in capital must, of course, be final in order to have the desired effect. It is also important to ensure that the owners will not find the time to violate the decision by transferring assets outside of the bank. The appointment of a temporary administrator with a mandate to sign off any large payment or asset transfer may thus be a useful tool.

The next step, to ensure that the state does not have to bear the creditors’ losses, is more difficult, however. The reason for this is that the banks’ funding consists to a large degree of borrowing with a short duration, which in turbulent times may tend to disappear.

There are sophisticated regimes under development…
It is possible to resort to more or less sophisticated solutions to deal with the problem.
One means is to fix a system that enables the authorities to take control over the bank and to write down not only capital but also claims in an orderly manner, while extending a guarantee to the new financing needed when the bank is under public control. The payments needed to maintain the stability of the financial system would be effectuated, while other payments could be temporarily stopped.

A draft model for such a system was presented by the Banking Law Committee in Sweden in 2000 and the proposal is currently being considered by the Swedish Government. As I understand it, there is also work being done in other countries to bring about solutions that would mean that the authorities could claim in all credibility that it was possible to reduce claims even in more complex cases, where there was a risk to financial stability. Ideas that have been put forward include immediately being able to impose a partial payment stop on certain claims, haircuts, in connection with the authorities taking control over the bank and transferring the assets to a temporary bank, known as a bridge bank.

To be effective, all these proposals make considerable demands for having IT systems in place that can quickly distinguish different types of claims. A number of other preparations are also required, for instance to be able to transfer contracts. It is thus not enough for the legal structure to be in place; there must also be a practical solution that quite simply does not provide any incentives to “push one’s way to the front of the queue”.

I welcome these efforts to strengthen the frameworks and look forward to seeing the results.

But I believe that one should have realistic expectations of the possibility to write down claims in a shaky market situation. The experiences from the Swedish banking crisis and also the Northern Rock case show us that in such situations it may be necessary to have an extensive guarantee of the banks’ obligations to stop a bank run or secure continued access to the international capital markets.

…and a simpler way which has proven its worth in real crisis situations
I would therefore recommend also introducing regulations that give the government the right to take over the ownership of a bank in a certain situation. If this is done in the right way, it will not only deal with the requirement that the shareholders should be held responsible, it will also minimise the risk that the government will need to bear costs that the creditors should bear.
For this to work, the government should have the right to take over the shares before the capital is completely wiped out. There are at least two reasons for this. First, there is a risk that the owners will “gamble for resurrection” when the capital is low. Second, it is almost always the case that the situation proves to be much worse than one thought. When the capital is at zero on paper, there may in actual fact be a large deficit.

It is not without controversy that the government is given the right to nationalise a bank. It may be particularly sensitive in countries that have a recent history of state ownership. But I believe that the possibility for compulsory acquisition of the shares is a proper tool. Again, the owners enter into the business knowing that a serious breach of the supervisory regulations should result in the revocation of the licence and the loss of control over the bank. The fact that an expropriation may be an alternative to the revocation of the licence in such a situation when financial stability is at risk is important to restore a proper balance of powers. As long as these powers of the state are clear beforehand, this should be acceptable to the shareholders.

Returning again to the Swedish crisis, I believe that it was a decisive factor that the banks realised early on that the government support measures were no free lunch. When this was made clear, the owners had strong motives to recapitalise the banks themselves.

Of course, the idea should not be to end up with a state owned banking system; equally important is to have an exit strategy.

Regardless of how the system for managing problems is constructed, it is also important that there is finality in the intervention. There are several examples of what can go wrong otherwise. Any errors should be corrected by means of damages, rather than reversing the decision.

There should be many tools
If there is a system in place that means there is a credible threat that the owners will lose their influence in a situation where the bank is in serious violation of the regulations, then there is scope to try out alternative methods of solving problems in individual cases. It is desirable that the authorities should have a broad range of tools at their disposal. It is reasonable that the deposit guarantee or the government can take part in a voluntary restructuring where the owners contribute capital, or that the government can issue guarantees in the case of a private takeover. Important is that owners and creditors should not be able to gamble on such solutions always being the cheapest option for the government.

Cooperative banks may be converted into a limited liability company
One special issue is how to deal with cooperative banks and savings banks. These are often small, but it cannot be excluded that problems in such banks can have systemic implications. It is not realistic to believe that a savings bank facing major problems can be restructured in its original legal form. To solve the problems, the alternative to winding up the bank and selling the assets is to convert the bank into a public limited company. We saw some examples of this, too, in the Swedish crisis, where some savings banks and other cooperative financial institutions had to be restructured.

Accounting issues need to be sorted out
When one gets to the point where the government takes control, there are also more practical problems to deal with. This is a stage where the assets must be inventoried and where bad loans may need to be transferred to special asset management companies. One important aspect with regard to the feasibility of a quick and efficient solution is therefore the existence of clear regulations for evaluating assets and liabilities in a situation where the government intervenes.

One particular question is how to deal with a situation where the government provides support for existing operations, for instance by injecting capital or providing a guarantee. The purpose of the intervention is to save the functions and not the bank as such. However, this action can have a positive effect on the value of the bank. In principle, it is reasonable that the value should reflect the value of the operations without reference to potential government support measures.

Additional legal issues
So if all this has been put in place, is one home and dry? At the national level, yes. On condition of course that a number of basic conditions are met. The most fundamental requirement is that there is an effective legal system. It is not possible to close one’s eyes to the corruption in the legal apparatus in some countries, making it almost impossible to establish that a bank has gone bankrupt in such countries. If this is the case, the type of insolvency framework has little importance. The importance of reliable property rights and registers and the possibility to seize collateral cannot be emphasised strongly enough. Finally, it is of course also important that the financial supervisory authorities dare to take difficult decisions. The people who work with these issues should therefore have sufficient legal protection.

International contagion effects and the need for a code of conduct
The international dimension has become more and more important. The financial systems are interlinked and there is a considerable risk of contagion effects. Many banks are also a part of international groups with a more direct financial link between parts of the group. All of this of course creates a need to discuss how to manage banks in distress at an international level.
The EU, with a single market for financial services and common regulations, has good reason to go further. Much has been done in the area of coordination, with a number of documents on how to coordinate in a crisis now in place. But I think that we could go further in the direction of setting up minimum requirements for a crisis resolution framework for banks and other institutions covered by the financial safety net. Here one can draw some parallels to the work on fiscal policy frameworks, where the EU has come a long way in terms of coordination and codes of conduct in the event of macroeconomic crises.

Where do we go from here?
It is time to sum up. My message is that it is important for us to get in place an efficient regulatory framework for weak banks. A framework which combines the need to maintain stability and minimise costs to the public sector in the short term with the requirement for long-term market discipline.

Returning to the health-care metaphor I started out with, we need to be able first of all to deal with the ordinary flus and broken bones of the financial system. But we also need to be prepared to deal with a more serious event. A possible outbreak of avian flu in the financial system.

I have tried to outline the broad lines of such a framework. In short, the following is needed:
- An insolvency regime which ensures that banks can be effectively closed at a low cost to the deposit guarantee system
- An effective depositor protection with a quick pay-out to depositors and a credible financing
- The ability to take special measures, such as guarantees, to avoid the failure of a bank threatening the financial system in its entirety
- A broad tool box to be able to enforce, also in those exceptional cases, the principle that the state will not bear the losses of shareholders and creditors, where it should be possible to tie conditions to any guarantee issued and with the option of temporary nationalisation as a last resort.

But drawing up a broad structure is perhaps the easiest part. After that, it is necessary to find a concrete legal framework. In the end, this responsibility falls on our governments. A few countries seem to have much of this in place already. And some have come quite far in this work and devote quite a lot of resources to the project. Those that have not I think are well advised to get started. It is just never a good idea to set up the fire brigade first when the city is on fire.

But it is a delicate task trying to find an efficient and legally secure system which will also work cross border. I therefore see an important role for you who are lawyers specialising in insolvency and restructuring in a much needed international debate on how best to strengthen the insolvency regime for banks and deposit taking institutions. This is an area where all of the good forces need to combine.

As long as we all remember to look up every now and then to check that the individual patches make for a nice quilt, I have no doubt that we will be able to come up with the necessary improvements.

With this rather positive outlook I end my remarks. I thank you very much for your attention and I wish you a successful conference.