Let me begin by thanking you for the invitation to come hear and speak today. It is a great honour to be asked to give a speech within the framework of this prestigious lecture series.
On many occasions I – and several of my colleagues at the Riksbank – have described the financial crisis in Sweden at the beginning of the 1990s, both here in Colombo and in other contexts. That is because this crisis made a considerable impression on the Swedish economy, resulting in low – and during a number of years negative – growth and high unemployment. However it also made an impression in terms of making us consider what basic conditions were necessary to ensure that a financial system functions smoothly.
There were many interacting factors behind the crisis experienced in Sweden. Mistakes made by bank management, by the supervisory authority and by politicians made the situation worse for the real economy. These mistakes were probably due to some extent to the financial markets having been regulated for almost half a century and so this affected the financial agents' way of thinking and perception of reality.
I intend today to describe in broad terms the design of, the background to and the effects of the Swedish regulations and then go on to discuss the deregulation process. After this I shall point out which factors led to the crisis and what it entailed for Sweden. This is because I believe that our experiences in Sweden are in many respects relevant to other countries. Although conditions tend to differ somewhat between countries, financial crises usually have similar causes and similar results. For example, comparisons have been made between the crises in South East Asia at the end of the 1990s and the crisis in Sweden. It appears that at least three quarters of the crisis causes were the same, including the excessively rapid credit boom, insufficient credit assessment and overheating in the economy. The sequence of events for the actual crisis was also similar; with large currency outflows, rapidly increasing loan losses and insolvent banks. I therefore hope you will have the patience to listen to my account of the Swedish experiences. I believe you will be surprised at how many similarities there are!
The Swedish regulations
During the period between the Second World War and up to the 1980s, the Swedish financial system was subject to a number of regulations. I shall mention some of the most important ones.
Credit regulation was one of these. Around half of the Swedish banks' lending must be directed to the Government and provided at a low interest rate and with a long duration. This was to enable the Government to finance ambitious housing construction programmes and budget deficits that arose as a result of investment and social welfare measures.
The banks themselves could decide who received the other half of their lending. However, the rate at which the lending could increase was determined by the Riksbank through “credit ceilings”. The idea was to keep credit expansion in line with economic developments and thus prevent inflation from accelerating. Moreover, the banks were encouraged to give priority to corporate lending so that the investment rate would not be threatened. Most of the loans to the private sector were made to large, international Swedish corporations.
However, it was not only the banks’ credit volumes that were regulated; the price of the credit was also regulated. The Riksbank determined both the deposit and lending rates. The charges for various bank services were also regulated. This was intended to prevent unhealthy competition and protect account-holders and borrowers.
The banks had to obtain permission to open branches in other towns in Sweden. This was to avoid too many banks establishing in one area so that competition became too stiff and profitability poor.
In addition to the domestic regulations, there were also a number of regulations regarding cross-border capital flows. These were more or less entirely regulated. Companies and individuals were only permitted to invest if they could prove the financial necessity of the investment. Foreigners were allowed to invest in Sweden without a permit, but these investments might not be financed in the Swedish market. Portfolio investment abroad was regarded as more transient than direct investment and it was therefore more difficult for a Swedish company to obtain permission for this in another country.
Companies that earned money abroad were obliged to bring it home to Sweden as soon as possible and convert it to Swedish krona.
Swedish banks required permits to establish subsidiaries abroad and were on the whole not allowed to open branches abroad. This was because it was easier to control capital movements between parent and subsidiary than between branches, as the latter were not different legal entities. Prior to 1986, foreign banks were not allowed to establish in Sweden. After this, subsidiaries were permitted and later on also branches.
Swedes were in general not permitted to purchase shares in companies outside of Sweden. Those who were already in possession of foreign shares were allowed to exchange them for others or to sell their right to own shares to someone else if they sold their holdings.
Private persons were not allowed to take unlimited currency out of the country. Amounts above the regulated limit – which was low – required permission from the Riksbank. It was necessary to show evidence that the money would be used for personal aims and not for investment or other financial purposes. Private persons were not allowed to hold accounts abroad. Of course, considerable resources were required to administer all of these regulations.
My description here shows clearly how thoroughly regulated the Swedish financial system was. Furthermore, the regulations applied over a long time. They were for a good purpose. The aim was to create stable markets for credit and foreign currency, secure and sound banks and also to achieve redistribution of wealth effects through prioritising economic policy measures. However, the result was the removal of necessary incentives for the banks to develop their operations in a professional manner and it also reduced the opportunities to compete, which prevented new banks from entering the market. Thus, short-term stability was attained at the cost of the build-up of long-term tensions, which would gradually result in a crisis.
The emergence of a market economy…
The idea behind the regulations was good, but they led to the central government having too much control over the banks’ operations. This put the banks’ functions out of operation.
The banks’ most important task is to borrow money from the general public and then lend it to companies who wish to invest. They must therefore be able to assess the creditworthiness of their borrowers. However, the regulations that had prevailed for almost forty years meant that the banks had not needed to develop thorough credit assessment competence. The central government had controlled their risk-taking. The limits to lending growth also meant that they could choose to only provide credit to the customers with the highest credit ratings. Loan losses were thus small. In addition, there was little competition and little incentive to develop new services. The Swedish banks had too many employees and lived in an environment that must with hindsight be regarded as sheltered. After enduring the crisis, the number of employees had been reduced by one third, but at the same time turnover had increased.
However, during the 1980s the regulation structure was slowly eroded. The banks’ balance sheets were full of government bonds that they had been forced to buy at non-market prices, and the credit flows began to move increasingly towards the unregulated sector outside of the banks. Developments in the international financial markets made the currency regulation increasingly difficult – and pointless – to maintain. It was not so much that there was a wish for deregulation; more that developments made it necessary. No one realised that this would entail a rapid and well-needed improvement in the efficiency of the banking system. Nor that the method by which it occurred would lead to a major financial crisis.
During this period a Swedish money market began to emerge. As the banks could not and did not want to buy more treasury papers the central government was forced to consider alternative methods of financing the central government debt. They began to issue first short-term and then longer term bonds through regular auctions and these were purchased by pension funds and insurance companies. The banks became mediators in the market – primary dealers – rather than final investors.
…and a financial crisis
How did the banks react to the new deregulated environment? Now they could set their own prices and grant loans as they wished. One might say that the bank management saw the opportunities more clearly than the risks. They wanted to obtain market shares and increase their income. The general public’s pent-up demand for credit was substantial when the credit regulations were abolished and high inflation made it more profitable to borrow than to save. The 1980s were also a period of strong economic growth, which contributed to the credit demand enduring even after the earlier pent-up need had been satisfied.
There was thus a rapid credit boom over a number of years. Between 1985 and 1990 the general public’s loans from banks increased from 75% to 115% of GDP. A substantial part of the borrowed money was used to buy property, which led to property prices being pushed up beyond what was essentially justified.
In this situation it is of course important to use monetary policy and fiscal policy to tighten the economy. However, at that time Sweden had a fixed exchange rate pegged to a basked of international currencies, primarily the US dollar. Monetary policy – the interest rate – had to balance the short-term capital flows and could not be used to influence domestic demand. Nor was fiscal policy tightened to cool down the overheated economy.
High inflation and rapidly rising wages meant that Swedish competitiveness was at a low level when international economic activity began to weaken in the early 1990s. At the same time, the Government made it clear that the fixed exchange rate should be defended at any cost. It was an untenable situation. Gradually, confidence in the Swedish krona deteriorated and this forced the Riksbank to raise interest rates to counteract the capital outflows. As inflation at the same time fell substantially – from 10% to 2% within the course of two years – real interest rates also rose substantially, from 1.5% to 8% between 1991 and 1992. The Swedish economy experienced a real interest rate shock.
This led to the property bubble bursting, which was a serious problem for the banks, as a large part of their lending had property as collateral. In some areas property prices fell by up to 60 per cent over an 18-month period.
During autumn 1992 confidence in the Swedish krona declined further. Despite the Riksbank setting the interest rate at 500 per cent over a weekend, it proved impossible to stop the speculative capital flows and on 19 November 1992 Sweden was forced to abandon the fixed exchange rate regime. This was a further blow to the banks. Many of the borrowers who had borrowed in foreign currency had their income or assets in Swedish krona and when the krona fell they experienced problems in paying their loans.
Confidence in the Swedish banks declined dramatically and they had problems in continuing to finance their short-term loans on the international interbank market. The bank crisis was upon us. To prevent the system from collapsing, the central government was forced to rapidly intervene with a general, overall blanket guarantee that promised that all of the Swedish banks would meet all of their obligations, existing and future, towards all types of lender.
All in all, the bad loans for the seven largest banks, when they were finally mapped out, amounted to 12 per cent of GDP. Six of these banks risked falling below the minimum limit for their capital buffers and therefore needed more capital from their owners or from the central government. An important principle in managing this crisis was that banks that were not viable would be closed down and their owners would lose capital. One bank was closed down and its assets were transferred to another bank, while the original bank was declared bankrupt.
The effects of the crisis
The effects of the crisis were considerable, not only for the banks but also for society as a whole. Growth fell drastically and was negative for three years in a row. Unemployment rose from just under 2 per cent to over 8 per cent between 1990 and 1993. However, the effects were also substantial with regard to the financial system itself.
As I see it, there are two reasons why the financial crisis in Sweden developed as it did and was so damaging. Firstly, there was a lack of the necessary competence for managing credit in a responsible manner, both in the banks and in the supervision of the banks – largely due to the long period of regulation. Secondly, monetary policy and fiscal policy were conducted in a way that reinforced, rather than moderated, developments.
I shall begin with the banks. Insight into the risks entailed in the new competitive situation was insufficient. The banks lacked the systems and processes needed to identify and measure credit risks, as well as to manage and prevent them. As you know, there have been rapid developments internationally with regard to awareness of these issues and knowledge of instruments to measure and manage risk. Even smaller banks now have comprehensive systems for assessing borrowers and can thereby avoid taking greater risks than they can afford.
If the banks analysed credit risks at all then, they only looked at the risks from an individual company. They did not analyse risks in entire sectors or those related to macroeconomic developments. To a great extent, the banks’ credit assessment was also focussed on the value of the collateral rather than the borrowers’ ability to pay. With hindsight we can observe that not even collateral was analysed in a satisfactory manner. Sound international banking practice today says that credit should only be granted on the basis of the borrower’s ability to pay and the financial strength of the project. Collateral should be required, but good collateral can never outweigh a borrower with a poor credit rating or an uncertain project. The bank must also make a realistic assessment of the value of collateral. For example, it is common to provide the borrower’s own premises as collateral. However, when a borrower goes bankrupt the value of the premises often declines substantially, as it is difficult and costly to adapt them to another user.
The banks accepted high concentrations, not merely in lending to individual borrowers, but also credit granting to individual sectors and geographical areas. For example, the banks’ loans to the property sector, or with property as collateral, accounted for more than 60 per cent of all loan losses. This is a concentration risk that contravenes all sound risk management and is currently prohibited.
However, it is easy to say all this with the benefit of hindsight. At the end of the 1980s, the banks had substantial profitability and low loan losses. Economic activity was strong. I believe that all of us active in the financial system then would have found it hard to believe that a crisis was impending.
Nevertheless, the Swedish banks had one important piece of the puzzle already in place prior to the crisis. They were well capitalised and had built up capital reserves that corresponded to, or exceeded, the statutory level of 8 per cent. This created some resilience and to a great extent reduced society’s direct costs when the crisis occurred, as the bank owners’ money was lost first. In some countries that have experienced bank crises, the central government has intervened and saved the bank owners’ capital contributions. In my opinion, this is both unfortunate and inappropriate and creates the wrong incentives for the future.
Like the banks, the supervisory authorities and legislators lacked insight into what risks were actually associated with banking activities in a competitive environment. This meant that the supervision remained largely as it had been before, with unchanged resources, despite a drastic change in conditions. There was neither the competence nor the instruments needed to assess what risks were arising within the banks. The main emphasis was placed on the banks reporting their positions correctly rather than on analysing the underlying risks. There were few economists within the Swedish Financial Supervisory Authority, but a lot of legal advisers. What was really needed was a suitable mixture of the two; and also other specialists in various forms of banking activities.
Our experiences of the crisis mean that I cannot emphasise strongly enough the importance of having carefully-drawn up legislation in place that can provide effective support for the supervision to be conducted. As well as support for potential crisis management. The legislation should provide the supervisory authority with obligations and powers of authority to intervene at an early stage against institutions that do not meet the requirements set, to avoid problems in one institution becoming magnified and spreading to others. The intervention by the authorities must be predictable and objective, so that the banks are not forced to operate in uncertainty. The legal rights of the parties in the financial market are important, at the same time as the authorities must be able to operate without the threat of legal countermeasures by the banks in the event of an intervention. The supervisory authority therefore needs to have adequate resources, the right competence and operational independence.
There need to be clear rules as to how the authority responsible can take over and close a credit institution. This was not the case in Sweden at the time of the crisis, which of course delayed the rescue work. Creating confidence in the banking system – which does not become unconditional confidence if the central government acts as final guarantor – also requires a deposit insurance that clearly indicates how much of the customers' deposits the government is prepared to guarantee. The legislation must also cover more than just the banks – other institutions that receive deposits from the general public must also be under supervision.
As I said, deregulation was necessary – it was rather the incapacity of the banks and the supervisory authorities to manage a competitive situation that contributed to the crisis. However, the economic policy conducted also contributed. Monetary and fiscal policy were not used to slow down developments at the end of the 1980s, which laid the foundations not only for increased borrowing, but also for the currency crisis. The combination of a fixed exchange rate and large budget deficit was devastating. Incentives for increased saving were not actually introduced until the crisis was actually upon us and then they only helped reinforce the decline.
It may seem something of a paradox that international organisations, such as the Basel Committee, are now implementing a number of new regulations – such as Basel II – despite the fact that the old regulations had negative effects and were therefore abolished. However, these are a completely different type of regulation. The old regulations were aimed at prohibition and other strict controls, while modern regulations give the banks plenty of flexibility to offer their services as long as they meet certain fundamental conditions. For instance, Basel II allows the banks to determine their own risk-taking to a great extent, as long as they have adequate risk management and capital reserves.
The experiences of regulation and the crisis also show clearly how important it is that the central bank can make independent decisions regarding the interest rate according to objectives set by parliament, for instance, to maintain inflation or the exchange rate at a stable level. If there are no guidelines for the interest rate, or if the parties that determine the interest rate have differing interests, it will often be set too low for the purpose of benefiting borrowers, investment and central government consumption. This was the case in Sweden during the years of regulation. Apart from leading to higher inflation and pressure on the exchange rate, exaggeratedly low interest rates also have a negative effect on the financial sector. If the real interest rate is low, even bad investments appear to be profitable and the banks' credit assessment process will not function properly. A large number of borrowers will then experience unsatisfactory profitability, particularly if the interest rate then increases slightly, and will be unable to pay their loans.
The bad experiences from the time before and during the crisis led to the Riksbank later being given a very independent role to determine the key interest rate within the framework for monetary policy. Parliament set the objective for what should be achieved, namely price stability. The Riksbank then specified this as an inflation target, whereby the inflation rate would normally be within an interval of 1 to 3 per cent a year. Interest rate decisions are made by the Executive Board of the Riksbank, which is intended to ensure that there can be no external influence. The predictable and independent interest rate policy means that the banks and other financial agents have been given a firm foundation on which to base their risk assessments, which reduces their risks of suffering unexpected losses.
Avoiding being wise after the event
It is dangerous to use regulations to limit competition in the banking sector. Almost forty years of regulations had led our banks to almost entirely lose one field of competence they must have to accomplish their task to society – to assess credit risks and allocate credit where it can have the most positive effect on economic developments.
When we were in the midst of the build-up phase of the crisis, there were few – if any – who had the capacity to see the consequences of the problems we were facing. That the banks had lost their ability to assess credit risks and were thus not functioning as they should. That those who supervised the banks did not have this competence either, nor the ability to see beyond the tables and the formal reporting. That the stabilisation policy could not manage to bring the economy back to a long-term sustainable economic development, as monetary policy was tied to the fixed exchange rate and fiscal policy was overly expansionary. Quite simply, that risks had been built up in several areas. At that time there was no authority that had responsibility for overseeing whether financial stability was in danger.
I certainly do not wish you a bank crisis; this is a story that is much better read than experienced. But there are some elements in the financial system and the financial developments here in Sri Lanka that give cause for consideration. This applies in particular to the macroeconomic picture. I hope that my presentation will give you reason to reflect on these problems, so that they do not grow and become unmanageable. We have already made all the mistakes in Sweden. It would be quite unnecessary to make them again.
Thank you!