Nyberg: Financial stability surveillance, Some Swedish experience
OBS! Talet finns bara på engelska.
Introduction
First of all I want to express my thanks to the Bank of Indonesia for inviting me to this conference. Financial stability - or the lack of it - is a very topical subject in the light of events in the past decades. Furthermore, financial stability issues seem to be strikingly similar between countries, whether they are big or small. It makes good sense, therefore, to have different central banks talk about their experiences. And, as always, the best experience comes from mistakes made. With this in mind, I shall tell you a little of how we in Sweden look at stability surveillance today.
First, some background. Systemic stability came under serious threat during the Swedish banking crisis at the beginning of the 1990s. As was later the case in Indonesia, the risk of severe financial instability led to government intervention. A decision at very short notice was required of Sweden's government and parliament in the autumn of 1992: Should they choose to support Swedish banks financially without really knowing how expensive the support might have to be, or should they risk a financial crisis with unforeseeable consequences for the Swedish economy? The decision that was taken was the only reasonable option - to rescue the Swedish banks. The government issued a general guarantee for the banks' liabilities and set up a special Bank Support Authority to cope with the banks that were in serious difficulties. The government, explicitly supported by the opposition, acted just as vigorously and promptly as the situation required. But this did entail making a decision - on the basis of insufficient information about the causes and depth of the crisis - that could have burdened the government budget with at least 10 billion US dollars. Having to take such important decisions at short notice and without adequate information about the financial system is an experience we do not wish to repeat.
The Riksbank is accountable to the Swedish parliament with regard to keeping a stable price level - our inflation target - but also with regard to promoting financial stability and a safe and efficient payment system. After the crises there was room for some self-criticism in this respect. In practice, we had not supported our stability target with any systematic analytical work on overall stability. To address this, we started to build up knowledge and analytical capability in the field of stability. By publishing a Financial Stability Report in 1997, we were among the first central banks to issue a periodic public account of our appraisal of financial system stability. Its example has been followed to date by about thirty other countries, including Indonesia.
Today, interest in financial stability is well established. The crises during the 1990s revealed serious flaws not only in macroeconomic management but also, just as importantly, in the structure, regulation and operation of the financial system. Most major financial crises in various countries around the world emanated from overriding economic or financial problems rather than bank-specific issues. If these overriding problems had been identified at an earlier stage, the crises might have been prevented or at least their effects might have been reduced.
Financial stability work
Banks are the focus of all work on financial stability. This is because banks are the institutions that ensure the functioning of the payment system. We know from experience that if individuals and companies are unable to make ordinary payments, or if the credit function of society is disrupted, this may have disastrous effects on the economy. Insurance companies, finance companies, security brokers and other firms may also be important in society, but they are not critical to the system in the same way as banks.
But banks are not only critical for stability. Their balance sheets also contain a built-in source of instability. Banks' funding is generally short term in nature and can quickly disappear, while their assets have long maturities and cannot be realised as quickly. Should depositors and financiers lose confidence in a bank, it could face an acute liquidity crisis. Moreover, problems in one bank can spread to other banks. Banks have liabilities to one another due to loans and securities trading, or because they participate in the payment system. Contagion effects can also arise if other economic players suspect that there are connections between the banks. Banks can then encounter difficulties, even if the original suspicions were completely unfounded.
Given this, the work on financial stability may be undertaken along three main strands. The first concerns the rules and regulations that set the bounds for the operations of financial institutions, notably banks. The second is the continuous surveillance of the system that is performed by the supervisory authority as well as by the central bank. Thirdly, there is the management of crises, since, unfortunately, we must assume that crises may occur and we must be prepared to handle them. I will discuss these three strands in turn, concentrating on the last two.
1 Regulatory framework
First, of course, financial stability rests on a legal structure, which establishes the bounds for the operations of financial institutions. The parliament is responsible for legislation on financial operations, while the supervisory authority issues more detailed instructions. In many countries, including Sweden, the experiences of the 1990s have paved the way for a thorough review of the legal structure of the financial sector. In this context, it is worth noting that laws and regulations today to a great extent are developed through international negotiations such as in the Basel Committee and, for us in Europe, the EU.
The role of the central bank in this process is the comfortable one of a knowledgeable observer, posing relevant questions in working groups and committees and pointing out inadequacies in suggested solutions. There is much to say about this process, but I shall have to leave it for a different occasion.
2 Day-to-day oversight
The second strand of financial stability work is the day-to-day oversight.
This may be described as resting on three pillars:
- the first pillar is the supervision of individual banks,
- the second pillar is the surveillance of systemic stability,
- and the third pillar is payment system oversight.
The first pillar is clearly the responsibility of the supervisors, whether they are a separate FSA or, as in many countries, a department within a central bank. The supervisors collect data from the banks, analyse these data and make on-site inspections. Prudential supervision performed by a competent authority must be the basis for all day-to-day oversight.
The second pillar is the joint responsibility of the FSA and central bank. The division of responsibility is really one of perspective. While the FSA bases its work on individual institutions, the central bank looks at the banking system as a whole and attempts to assess the risks that could arise in it both in the short and the long run. Of course, there is overlap in the tasks of the FSA and the central bank, but it is often fruitful to approach a problem from somewhat different angles. Details are important, but looking at details you sometimes miss the overall picture. This, at least, is one conclusion we drew in Sweden from the experiences of the 1990s.
The third pillar of oversight is the supervision of the financial infrastructure. This is a relatively new task derived from the complicated technical and legal systems that handle all financial transactions today. The infrastructure is of particular importance to the central bank, since we would most probably have to provide emergency liquidity assistance if anything went wrong. The infrastructure is dominated by the systems for clearing and settlement of payments. In Sweden, like in many other countries, we manage the large-value payment system in-house at the central bank.
But the infrastructure also includes the stock exchange, the derivatives exchange, the central securities depositary (CSD) and connected trading systems. Well-designed systems reduce the risk of contagion from the financial markets or between institutions. They also reduce the risk of the infrastructure itself causing problems through, for instance, operational disturbances.
The Financial Stability Report
In Sweden, the main tool for addressing the oversight of systemic stability is the report on Financial Stability. The Report, in Sweden as well as in other countries, basically aims to provide a picture of the banks' ability to stand up to any disruptions. Factors which might influence banks simultaneously are in focus, for instance the emergence and bursting of financial bubbles, excessive debt leverage in a certain economic sector or structural changes leading to reduced revenues or increased risks. The purpose is to identify conceivable risks and present an analysis that will influence the behaviour of market participants.
A first and very relevant question is whether a stability report really would help to prevent a crisis. Will the financial sector read and listen to what the central bank has to say? Will this kind of moral suasion be sufficient? Could we have written a report in the late 1980s that would have made the banks see what they did not see? We do not know. But at least as long as managements have a memory of the last crisis, they are likely to listen. And if the crisis develops slowly, there are certainly things to be done. I shall return to this question when I talk about crisis management.
Let me now turn to the main elements of a stability report as I see them. First, we look at the banks' borrowers, because as long as they are healthy the banks usually are as well.
Borrowers are analysed in terms of the corporate and household sectors in general, with particular reference to levels of debt and the future ability to pay. The real-estate sector is considered separately because lending to this sector is in most countries a major part of banks' total lending. The prices of commercial properties and private housing are also analysed not least because they reflect the value of the major part of the collateral that is pledged for loans from banks and mortgage institutions. Insofar as your banks conduct activities in foreign countries, stability surveillance also needs to encompass developments in these countries to ensure that negative developments will not destabilise your banks.
Second, the major banking groups are analysed in various respects. Profitability trends illustrate the ability of the banks to build up reserves for future unexpected losses. Moreover, there is always the strategic risk that a bank with weak profitability will be tempted to try out new, bold but risky lines of business in order to generate a higher return on equity.
Our analysis of the banks' credit portfolios provides an indication of how loan losses are likely to develop in the coming year or two. Another aspect of the analysis relates to banks' liabilities. Traditional banking involves accepting deposits and using the funds for making loans. People's growing propensity to save in other instruments than bank accounts has forced the banks to finance a major part of the growth in credit by borrowing, mostly short-term funds, in the capital market. The structure of bank funding provides an indication of the risks in bank liquidity. Monitoring this is perhaps the most difficult part since the liquidity situation is liable to change very rapidly.
3 Crisis prevention
Turning to my last financial stability issue, that of crisis management, in my view this is mainly a question of being prepared. Or as prepared as possible, I should say, because how can you adequately prepare for the unknown?
There are basically two types of crisis: a slow crisis and a fast crisis.
By a fast crisis I mean single but severe events like fraud, major loan losses or general events that quickly lead to loss of market or creditor/depositor confidence. These cases call for quick action. Normally the central bank takes the lead because of its ability to provide emergency liquidity assistance and because of its access to updated information on the liquidity situation in the market. But close cooperation is required, especially with the supervisory authority, which has information on individual banks, and with the Ministry of Finance, if there is a need for solvency support. In my experience, there are many possibilities for such a process to fail. Misunderstood information, unclear leadership or conflicts of interest may delay the process until it is too late. The chances of solving a fast crisis increase enormously if all institutions involved have prepared for it, e.g. through crisis management drills. Such drills tend to uncover the shortcomings or conflicts that may prevent smooth cooperation when really needed.
A slow crisis, on the other hand, may be detected but not always prevented by financial oversight. In a slowly developing crisis there is time to prepare action, which may even include new legislation, for instance on bank resolution.
To detect such a crisis may sound easy, but is in practice very difficult. For instance, when does a gradual increase in asset prices actually turn into an unsustainable bubble and no longer constitute a reflection of fundamental values? Look at the housing markets in some countries around the world right now. Are there bubbles and, if so, where are they?
Another problem is timing. At what point should the central bank sound the warning bell? By acting too early, we might add to the volatility in the market place and hence worsen the situation. Remember that the central bank will be held accountable for the negative repercussions of pricking the bubble, not only for the potential but less tangible benefits of having done so. By acting too late, on the other hand, we do not fulfil our mandate and a full-blown crisis might not be prevented.
In addition to the problem of timing, there is the problem of selecting an appropriate response to the threat. The first line of defence is moral suasion, i.e. trying to convince the banks and other market participants to act in a proper fashion. The problem with moral suasion is that it may be difficult to calm a market that is rushing to new heights. Still, by publishing financial stability reports and issuing warnings to relevant parties, it should be possible to influence the public discussion. The second line of defence is prudential regulation. Discussions could be held with the supervisory authority and the Ministry of Finance on solutions, such as stricter formal regulations, calls for improved risk management systems or other conditions in order to strengthen banks. A good example is the Hong Kong Monetary Authority, which, in a situation of unrealistic property prices, used its power to gradually lower the loan-to-value ratio of property collateral in order to limit bank lending.
A third line of defence of course is the use of monetary policy, which in practice probably means raising interest rates to prick a bubble. This, in my view, should clearly be an option. But there are also some strong arguments against a central bank trying to respond to asset prices and credit expansion by changing its monetary policy stance. And there are certainly situations when it is unlikely to work. I shall have to leave this issue for now, however, or you will have to listen to me for another hour.
Conclusion
Let me conclude by saying that much has been done in the work on financial stability in the past years. Still, it is not hard to identify areas with room for improvement. The analysis of borrowers and other counterparties to banks in the home country, but also abroad, needs to be enhanced. Central banks need to acquire a better understanding of how the banks manage their liquidity, how they select their sources of funding and the effects this has on their resilience to disturbances. The analysis of stability also needs to be supplemented with methods for stress tests, that is, ways of analysing the sensitivity of banks to shocks. Another area that will prove challenging for both stability analysis and risk management preparedness in the future is the ongoing integration in Europe and, more generally, banks' tendency to set up cross-border establishments in many countries.
I see great scope for close international cooperation on these issues, both in terms of developing methods for analysis and for other aspects of stability surveillance and crisis management. I would like to thank the Bank of Indonesia once again for devoting an international conference to these important topics.
Thank you!
