Heikensten: Risk management and financial stability
The whole speech
Why is the Riksbank interested in risk issues?
Firstly, I should like to thank you for the invitation to come and speak at the Risk Management Forum.
The Riksbank's objectives are not limited to price stability; the Sveriges Riksbank Act also requires that we promote a safe and efficient payment system. We therefore have a natural interest in monitoring risks in the financial sector and that these are handled in a responsible, but also efficient, manner. As you know, this has not always been the case. At the end of the 1980s, risks built up in the Swedish banking system that were much greater than most people realised. The Riksbank was as unaware of the situation as others. However, when the crisis was over, we decided that we needed to improve our knowledge of the risks in the financial system and to closely follow their development. Later, we began to publish a twice-yearly report on financial stability, which focuses on the development of risks in the financial system. We were actually pioneers in this work and several central banks have since followed our example.
Naturally, our endeavour has not been to ensure that the risks are as small as possible in every aspect. On the contrary, one of the most important tasks with regard to the financial system is to facilitate risk-taking; to manage and redistribute financial risks. Methods for this have developed at an incredible rate over the past decade. The new possibilities for risk management have strongly contributed to increasing the efficiency of the financial sector and thereby the economy as a whole. However, it is important that the risks are managed carefully and that they are reasonable in relation to the capital allocated to cover them. They must therefore be clear and transparent to all involved. In this respect, the financial institutions and society as a whole have no fundamentally difference of interests. No one wishes to be surprised by risks they knew nothing about.
The Riksbank also works on financial risk issues in many international contexts, in the Basel Committee, the IMF and on committees within the EU and ECB. As the financial markets to a large degree operate across national borders, it is often necessary to establish joint regulations and principles at international level. Throughout, the important issue is to weigh stability concerns against efficiency aspects, to find regulations that make the risks transparent and limit them in relation to the available capital. This should also be achieved without excessively encroaching upon the efficiency and development of the markets.
My overall theme today is how the Riksbank views developments in the risk management field from the perspective of financial stability and efficiency. However, before I go into this further, I should like to provide some background to this by reminding you of a time when some risks were managed inadequately, which led to a serious systemic crisis, namely the Swedish banking crisis at the beginning of the 1990s.
The banking crisis from a risk management perspective
Taking a narrow view, one could say that the banking crisis was the result of the institutions and authorities lacking the capacity to assess and manage risks. The banks came from an environment that had been strictly-regulated for many decades and where their operations were so firmly controlled and restricted that a cursory awareness of risk management sufficed in day-to-day operations. The demand for credit was much greater than the supply, which meant that only those with the best credit ratings were allowed to borrow. When the regulations were gradually phased out during the 1980s, the banks and other financial institutions lacked the expert knowledge and organisation necessary to manage the new, increased risks, which were also more difficult to assess. As this curb no longer existed and the banks instead began to compete for market shares, they took on hasty and in many cases unhealthy risks. We were not the only country to experience this development. Over the past fifteen years, many countries around the world have suffered financial crises for similar reasons; as the result of financial deregulation, which has not been combined with the necessary improvements in risk management and supervision.
The authorities responsible for the institutions and for financial stability, mainly Finansinspektionen (the Swedish Financial Supervisory Authority) and the Riksbank, had not built up sufficient competence regarding the financial risks. Nor did the legislation suffice as a basis for the authorities to limit imprudent credit-granting. The oversight of the banks and other credit-granting bodies therefore became formalistic and fragmentary. We were not always looking in the right places.
In defence of the banks' management and the authorities it can be pointed out that risk assessment was made difficult for many different reasons. Several important factors counteracted a sound financial development, such as the high inflation rate, misleading tax incentives and central government subsidies. In addition, there was a lock-in effect in that currency regulation, which was in force until 1989, had kept our capital inside Sweden and contributed to a form of hothouse effect. The banks did not take these factors into account in their risk assessment, or not to a sufficient degree. They then suffered the consequences when the government made some decisions that were desirable in themselves, but which altered conditions drastically. Inflation was pushed down, the tax system was reformed and subsidies were erased.
Developments in the property market were a central component in the banking crisis, as the banks' credit-granting meant they were highly exposed to the property sector. Property prices rose at an unreasonable rate, regardless of how one measures them. They left the basis for normal pricing, i.e. the net return for the near future and instead reflected pure hopes of what a future buyer of a property might be prepared to pay. When the property bubble burst, prices plummeted, in some cases by over 60 per cent in around 18 months.
There were several weaknesses in risk management with regard to property loans:
- An incapacity (or unwillingness) to see the price trend as a speculation bubble. Some people did realise the dangers, but had trouble making themselves heard in the prevailing atmosphere.
- The emphasis on the loan collateral, i.e. the property itself, at the cost of an assessment of the borrower's capacity to repay.
- The fact that no one saw the substantial concentration of risk through exposures that were ultimately dependent on property price trends. It has been calculated that more than 60 per cent of the banks' loan losses during the crisis can be attributed to property loans.
- The bank's "self-delusion", which consisted of high-risk loans, i.e. loans against collateral of low priority, being granted by financial companies that were closely linked to the banks. It was among these companies that the financial crisis began. In other words, there was a lack of overall group perspective in the risk management.
As the rise and later the fall on the property market played such a central role in the banking crisis, it is worth a special mention. However, it was not the only risk factor the banks should have managed better. There were also inadequacies with regard to credit granting. Large loans were made to new customers, new sectors and new geographical areas. This type of lending strategy is not wrong in itself, but it requires that the credit-granter has done his homework and is aware of the possible risks. In some parts of the banking world, they talk about credit-granting according to the church tower principle, i.e. that loans should be primarily made to customers living and active close by. How this principle was applied in connection with investments in Atlanta or Brussels is not easy to see.
In addition, the banks contributed to their customers building up a comprehensive credit exposure in foreign currencies, regardless of whether or not they had matching income in their loan currency. As the krona exchange rate was high, some borrowers chose to take out loans in foreign currencies that had a lower nominal interest rate. The risks with this strategy became evident after the krona's rapid fall of almost 25 per cent in six months, calculated from when the krona was allowed to float in November 1992. When the banks' borrowers found that their debt amounts had increased by up to 25 per cent, calculated in kronor, many of them were unable to meet their obligations and the banks' loan losses increased further. These actions by the banks are actually quite remarkable. The speculation in continued price rises in Sweden was evidently completely contrary to the belief that it would be possible to maintain a fixed exchange rate in the long term.
With hindsight, I can observe that financial risk awareness in many parts of Sweden was low during the 1980s. Since then, however, there has been a rapid improvement.
From regulation to a market-based financial sector
The 1980s entailed a transfer from a severely regulated and controlled financial sector in Sweden - as in many other countries - to a deregulated and market-based sector. Deregulation created better conditions for the financial sector to provide an efficient contribution to macroeconomic growth, not least by freer granting of loans and other forms of financing and by managing and distribution of risks. However, to ensure this could be done without the type of problems that led to the banking crisis, the banks' risk management had to be improved. Over the past 30 years, developments have been very rapid with regard to measuring and managing certain risks and enabling the end-customer to choose risk level. The breakthrough for option assessment in the 1970s and the emergence of the swap market in the 1980s were followed by the development of credit risk management instruments in the 1990s, such as securitisation and credit risk derivatives. This is a development which is continuing in important areas, such as market risk and credit risk and related markets. The breakthrough in financial theories has contributed to this development and made it possible to create instruments for buying and selling risks.
When writing history, we must not forget IT developments. Without these it would not have been possible to rapidly calculate and update the values of instruments, model parameters and risks in individual cases or in portfolios. The emergence and development of markets for risk management instruments and even institutions' internal risk management would have been delayed.
The combination of new financial instruments and markets has led to better conditions for risk distribution, where those who are best-equipped and most willing to take on risks can do so, while those who want lower risk can reduce their exposure. Sometimes, risk distribution is seen as a zero-sum game, but this is incorrect. A financial agent may take on further risk if this covaries with other risks in such a way as to reduce the total risk in the agent's portfolio. This example can be stretched to cover society as a whole, where there were probably people and companies who had not achieved the risk portfolio that best reflected their willingness and ability to take on risk, but will now be better able to do so with the new instruments and markets.
My overall assessment is that there is no doubt that this development has contributed to the financial system's capacity to carry out its fundamental tasks, while its resistance to destabilising shocks has been reinforced. But has the development of the derivative markets and the often sophisticated derivative instruments really led to better risk management? I believe it has. Derivatives redistribute risks to other risk-takers. These risk-takers may, of course, be better-equipped or more poorly-equipped to manage risks than a bank. If the risk is realised, it may lead to large losses in other parts of the financial sector, for instance, the insurance companies. However, if the risks are really diversified to a large number of agents, this should on the whole be positive from a macroeconomic perspective. One could say that we will all bear a small part of the gain or loss on our future pensions, but we hopefully avoid financial and macroeconomic costs from banking crises of the type we experienced in Sweden at the beginning of the 1990s.
Given the significance of risk management for society and for financial stability, it is natural that the Riksbank should carefully follow developments in this field. During the era of regulations, we were the authority with the task of monitoring that the banks observed credit policy restrictions in their operations, but we now have a completely different direction. We closely follow - and sometimes even try to influence - how risks are managed in the financial sector. Risk management should be governed primarily by the needs of the customers, markets and intermediaries. It is only necessary for authorities to intervene with regulation and other means of control where there is a risk of market failure, i.e. when the interests of the institutions and the markets do not coincide with the interests of society as a whole - the lenders' short-term market-share philosophy at the end of the 1980s is an example of this. If means of control are needed, they should be designed as far as possible to lead to solutions that are compatible with the normal working of the institutions and markets.
Some problems
We can all see that interest in risk management has increased and that considerable resources are being invested here. The interest applies to both the production of new financial instruments, the emergence of new markets and the practical measuring and management of various risks. In addition, awareness is being stretched to cover new types of risk that did not previously receive the attention they deserved, such as operational risks, interest risks throughout the bank, liquidity risks and legal risks. This is essentially a positive development, but let me nevertheless conclude by pointing out some phenomena and situations, which indicate that the new developments have also brought about problems that need to be discussed and hopefully solved.
The best known example is perhaps the LTCM debacle. The hedge fund suffered a sudden decline in value, which led to a confidence crisis among customers and it had to be shored up with liquidity infusions of many billions of dollars from a group of banks. The Federal Reserve even assessed the situation to be a possible threat to the overall stability of large financial markets. How could this happen with two Nobel prize-winners responsible for the strategy and risk models, which should guarantee that they were designed according to the latest research? The weakness lay in the fact that these models, like all others, were based on a given probability of history repeating itself. However, in autumn 1998, reality in the form of various simultaneous political and financial events took a different turn than was usual. It should also be added, for accuracy, that LTCM also suffered because its risk strategy was not very transparent. Investors were unable to assess to what extent the fall in value was temporary or permanent and therefore preferred to recall their investment. The main conclusion is nevertheless evident. Don't rely too heavily on historically-based models.
A further potential problem is that risk models are standardised. The demands from the supervisory authorities that the models should meet a number of minimum criteria could lead to the management and assessment of the risks becoming more and more similar. There is a fear of a destabilising lemming effect, with all financial agents running in the same direction when new data is received. In addition, there are fears that the standardisation leads to certain categories of borrower experiencing more difficulty in obtaining loans in future. The possibility of going to a different bank to get a loan could be reduced if all banks apply largely the same credit assessment system. I believe that these worries are exaggerated, on the basis of studies made by the Riksbank and others that have indicated that different banks tend to assess individual customers differently, even when they use similar risk models. This could be because they have unique knowledge of the customers, or because their present loan portfolio favours accepting a particular customer. Although I am not excessively worried on this account, it is important to ensure that international and Swedish regulations do not lead to unnecessary standardisation.
Operational risks have come to the fore in recent years, partly through a number of events, September 11, of course, as well as others such as large-scale power failures, IT and telecommunications breakdowns and the Basel Committee's capital adequacy requirements. The development of methods to manage operational risks has not come very far. Attempts are currently being made to rectify this. An increasing number of bank managers - both in Sweden and abroad - are showing a positive attitude to the focus on operational risk management, as they see these risks as potentially large ones. From the Riksbank's stability perspective, I can only agree - events affecting practical operations in the financial system could threaten stability. I therefore consider it important that the banks continue to build up their risk analysis and their protection against risks in this field.
It is possible to view the development in itself as a risk in that it has created new, potent instruments and possibilities for speculation. If used incorrectly or carelessly, these instruments could destabilise the financial system or in other ways lead to inconvenience. It is therefore important that both the financial agents themselves and the authorities responsible in this field closely follow developments in both theoretical and practical work and are prepared to take action if they identify risks to stability. Of course, it is another matter in this context - as in others - that the costs of new methods of control must be weighed against the risks to be managed.
Regardless of the fact that there are now outstanding issues to be resolved to improve risk management, my overall conclusion is positive. The development underway with regard to identifying, measuring and managing risks provides possibilities to improve stability in the banks and other parts of the financial sector. We still have some way to go, but the work is proceeding and is presently in an intensive phase in both the financial institutions and the authorities concerned. The Riksbank intends to continue to play an active role in this work.
Thank you.
