Nyberg: Are the risks growing in the banking system?

I would like to thank you for inviting me to today's analyst meeting and giving me the opportunity to discuss financial stability issues with such a qualified audience.

I would like to start by briefly explaining the Riksbank's analysis of financial stability. In addition to the inflation target, the Riksbank has another primary task, namely that of being responsible for financial stability. These two important functions are closely linked. A stable financial system is important for price stability since financial imbalances can threaten the inflation target. Likewise, an inflation target policy contributes to stability in the financial system, as it has a levelling effect on cyclical fluctuations.

The Riksbank's analysis of financial stability can be interesting for analysts of bank and property shares, since we closely monitor the banking sector and property market within the scope of this analysis. Our perspective is somewhat different, however, since we are mainly interested in risks which can threaten the stability of the financial system and which often appear to be relatively improbable, while stock analysts attempt to assess the most likely developments in the market. However, the information about the markets is largely the same and analysts also need to consider the risks in the companies which they are covering.

After this introduction, I will then go on to discuss the strong credit expansion in Sweden and the measures that a central bank can take if it assesses that this development constitutes a potential threat to financial stability. I will conclude by looking at international developments, since the development in Sweden is by no means unique.

The comprehensive changes that have taken place in the financial sector and the financial crises that have occurred in various parts of the world have meant that the work on securing financial stability has become increasingly prioritised. Those countries which have undergone severe banking crises in modern times are now leading the way in terms of the work on financial stability. For example, Sweden was the first country in the world to regularly and openly publish special stability reports analysing the situation of the financial system. We would say that the initiative was received with mixed feelings by other central banks, some of which were concerned that these kinds of reports would cause problems rather than preventing them. However, the attitude of the Riksbank has been that it is through open and honest communication with the financial players that we can observe and remedy potential problems at an early stage. Despite some initial scepticism, several countries have now followed the Riksbank's example. Norway and the UK now publish stability reports and Denmark also recently issued its first stability report. Since the financial markets operate across national borders, the Riksbank works together closely with foreign central banks on these issues. For example, the Riksbank participates in several ECB committees related to stability issues and takes an active part in the ongoing work to reform the capital adequacy regulations.

It is also stipulated in law that one of the primary functions of the Riksbank is to promote a stable and efficient payment system. In order to fulfil this task, the Riksbank must closely follow developments in those financial institutions that are of central importance to the payment system, i.e. primarily the four major banks, which are each sufficiently large to cause problems for financial stability if they fail. The analysis focuses on three main areas. The first area concerns general developments in the banking sector – how the competitiveness and long-term profitability of the banks are developing. The second area examines credit risks related to the banks' lending to households and non-financial companies. The third area that is analysed is counterparty and settlement risks that arise in connection with the banks' trading and taking positions in the financial markets. The analysis is undertaken on an ongoing basis and is presented in the Financial Stability Report, which is issued twice a year.

I will mainly focus on the analysis of the banks' credit risks associated with lending to households and companies. This is a subject that has attracted growing attention as lending, especially to the household sector, has risen significantly. A certain increase in lending is normal in an economic upswing, when companies need to borrow in order to invest, while households borrow, for example, in order to meet rising house prices. In an economic boom, loan losses are also generally low, while they tend to be higher during a recession.

The increased level of indebtedness has also come under discussion. The question has been raised of whether households and companies today borrow to such an extent that this could lead to problems in the banking sector when the economic cycle turns downward. History shows us that the factor which has triggered banking crises in modern times has often been extensive loan losses caused by changes in the macroeconomic environment. This was also the case during the Swedish banking crisis at the beginning of the 1990s.

So what is the situation at present? Figures for April show that total lending to the general public is up 7.3 per cent compared with April 1999. The main growth is in household lending, which has risen by 8.8 per cent, while corporate lending has increased by 4.1 per cent.1) A perspective on these figures can be obtained if we consider them in relation to GDP. This gives a measure of whether lending is growing at a faster rate than the economy as a whole, which provides us with an indication of whether imbalances are being built up.

In terms of household lending, the situation is rather different. Lending to households has risen relatively strongly over the last two years and currently amounts to over 40 per cent of GDP. It is therefore approaching the levels we saw at the end of the 1980s. However, it should be stressed that households' interest expenses are currently lower due to the low interest rates. This rise in lending is also reflected in the household debt ratio, i.e. debts in relation to disposable income, which has risen and which now amounts to 104 per cent .

Households are borrowing to an increasing extent, at the same time as their economic situation has improved as a result of rising disposable incomes and falling unemployment. Households have traditionally borrowed mostly for housing, and the current increase in this type of lending is perhaps not so surprising in view of the fact that housing prices in metropolitan areas have risen relatively significantly. What is more remarkable is that lending for other purposes is increasing at a higher rate. It is difficult to see from the statistics exactly how the borrowed money is used, but lending both from banks and finance companies is now growing at a faster rate than lending from mortgage institutions. This has given rise to apprehensions that households are borrowing for consumption or in order to invest in shares, which could lead to rapid increased risks. In order to gain a better understanding of whether or not this is the case, the Financial Supervisory Authority has recently commenced an examination of the banks' lending to the household sector.

There are several reasons why property prices play a central role in any discussion of financial stability. In addition to the fact, as I mentioned earlier, that the credit demand among households is largely governed by the demand for and price level of property, property can also be used as collateral for loans. Another reason for monitoring the property sector is that a large part of the banks' lending is to property companies. These companies also accounted for a large part of the loan losses during the banking crisis in the early 1990s.

Property prices have risen recently, and in particular the prices of residential property in metropolitan regions. In contrast to the 1980s, we do not see the same broad price increase across the whole of Sweden, as was the case then. Likewise, the causes of the rising prices appear to be low interest rates, a high rate of new occupancies and low construction activity rather than speculation. This indicates that there are currently no major risks of a fall in residential property prices. However, there is every cause to observe developments carefully. New construction always takes place with a considerable time lag and experience shows that many projects are not ready and do not come onto the market until after a turning point has occurred, at which point things can quickly deteriorate.

With regard to commercial property, the development has been more moderate. Again, it is mainly in the metropolitan areas that prices have risen due to high demand. The level of new construction continues to be low and most premises today are already leased before building commences.

One way of analysing the extent to which speculative tendencies are inherent in property prices is to place the direct yield in relation to risk-free interest.2) At the end of the 1980s, the risk-free interest rate was considerably higher than the direct yield on properties, which suggests that people bought properties on the expectation of future price rises and/or significant rent increases . Today, the situation is different. Direct yield levels appear reasonable in relation to risk-free interest. With regard to apartment buildings in Stockholm, the direct yield is lower than the risk-free interest, but at the same time prices are affected by the fact that most property deals are made in connection with converting apartment buildings to tenant-owned apartments. For these type of transactions, direct yield cannot be used as a measure since the tenant-owners association does not have a required return comparable to that for an owner of a rental property.

The financial statements presented by property companies are also considerably stronger today than what they were ten years ago. The risks associated with lending to this sector by the banks appear to be considerably lower than then. Among listed property companies, it seems as though many consider that the level of indebtedness has become too low, considering the extensive interest in buying back shares within the sector. At the same time as buying back shares can lead to a higher return on equity, there is reason for companies and their credit granters to observe the risks associated with higher indebtedness. Above all, the companies can become extremely sensitive to changes in interest rates.

An overall assessment of the credit situation in Sweden thus suggests that there is greater cause to be pleased that lending is growing as an integral part of the current economic upturn than to worry about the risks which may accumulate. However, an integral part of the Riksbank's role is to remind about changes that can occur both rapidly and unexpectedly.

Getting back to the credit expansion in the household sector, the development we see in Sweden today is not unique. The American economy has been the subject of much recent discussion. A growing number of voices have warned that the risks have been built up into something of a financial bubble. The apprehensions derive from highly inflated asset values, low saving and increased indebtedness, i.e. in principle, a similar situation to the one we currently have in Sweden. Total lending by American banks increased in March by 10.1 per cent compared with March 1999 and lending in relation to GDP is currently at a historically high level . The lending statistics show that the primary increase is in lending for housing, due to rising property prices. A the same time as lending is increasing, saving is declining. Over the last five years, saving as a proportion of disposable income has fallen by nearly 5 percentage points and amounted to a low 2 per cent during the last quarter of 1999. Sweden has shown a similar development. During the last five years, the saving ratio has fallen by over 9 percentage points and is currently at 2 per cent. There are thus several signs of a development that is worrying from the view of financial stability. The rising debt ratio has without a doubt made households more vulnerable. However, as in Sweden, employment and households' financial assets have increased, which favours the households' ability to pay.

Another example is the Netherlands which, like Sweden, is normally characterised as a small, open economy. Even there, worrying signals similar to those in the Swedish and American economies can be perceived. The improved economic activity has led to falling unemployment and rising asset prices. During the last two years, total lending has risen at an annual rate of around 15 per cent. In the Netherlands too, a large part of this rise in lending can be explained by rising prices, especially house prices. Dutch house prices rose on average 15.8 per cent during 1999. As a comparison, it can be mentioned that Swedish house prices rose by approximately 8 per cent during the same period (prices for tenant-owned apartments rose to a considerably higher extent, however). In the Netherlands, household indebtedness in relation to GDP has traditionally been low, but has risen dramatically in recent years and today amounts to almost 45 per cent of GDP.

If we compare the developments in Sweden, the USA and the Netherlands , we see Dutch property prices have risen considerably faster than Swedish and American prices both in real and nominal terms. In this context, the development in Sweden appears to be less sensational.

An increasing level of indebtedness in the economy, combined with rising asset prices, involves greater vulnerability to changes in the macroeconomic environment, i.e. growing risks. The obvious question then, is what the authorities can do to prevent such a build-up of risk. The Riksbank discusses a number of possible measures in its latest Financial Stability Report. Firstly, the Riksbank tries to influence developments by communicating our views on what is happening publicly and in the form of a dialogue with the banks. The Financial Stability Report plays a central role in this context. The Riksbank can in certain situations provide a more comprehensive picture of the risks than the individual market players, as the latter attach greater importance to evaluating developments in their own portfolios.

If the Riksbank believes that an unhealthy build-up of risk is occurring and our assessment has not had the intended influence on the individual market players, there are other measures that the Riksbank can take. In extreme cases, quantitative regulations may be introduced, such as raised required capital adequacy ratios or a reserve requirement. Nor can an increase in the repo rate be ruled out if financial stability is threatened. However, these are more far-reaching measures and would only be used if the build-up of risk was assessed to constitute an acute problem. The assessment of the situation made by the Riksbank and other authorities will always be marred by a large measure of uncertainty, as it is extremely difficult to evaluate how serious the build-up of risk actually is. If drastic measures are to be taken, the assessment must be based on a strong foundation.

The development we can see today is, as I mentioned earlier, far from requiring any intervention. The Riksbank's assessment is that if the macroeconomic situation should worsen, it will lead to problems for individual households in meeting their payment obligations, but will hardly comprise a serious risk for banks in general and therefore not threaten financial stability either. In this context it is important to mention that the household sector normally accounts for a small percentage of loan losses, even during the banking crises. There is thus no reason in the current situation for the Riksbank to resort to special measures to intervene against the build-up of debt currently occurring in the household sector.

When discussing the banks' loan losses, I have alluded to the risk that households and companies will not be able to pay their loans. Loan losses can also arise when the banks cannot meet the obligations among themselves. These risks are usually called counterparty and settlement risks. This type of risk arises in the banks' trading and taking positions in the financial markets. As the major Swedish banks have extensive commitments to one another and also to foreign banks, problems in one bank can rapidly spread to the others and cause problems throughout the entire financial system. Such a problem need not be due to a bank being insolvent; it may be based on, for example, a failing computer system or similar. The Riksbank has on several occasions pointed out that these individual exposures are a growing problem from the point of view of stability. To overcome this problem, the Riksbank has begun a review of the regulations, together with the Swedish Financial Supervisory Authority.

In the current situation there are great expectations that the future will bring economic growth and good times. "The new economy" with a low inflation rate and high levels of employment and growth has gradually crossed the Atlantic to reach us. In such times of strong confidence in the future, consumption and indebtedness increase, while saving declines. We have seen this happen many times before. If the expectations are fulfilled, there is no problem. However, if the expectations race ahead of economic developments, asset prices and indebtedness can develop in a way that causes imbalances. If there is an economic downturn, households may find difficulty in meeting their payment obligations, which will lead to a rapid increase in saving, falling asset prices and a decline in demand. This is also something we have seen in other contexts.

While this development could give cause for concern, it is important not to paint an excessively negative picture. As the Riksbank has pointed out earlier, a decline in the macroeconomic situation would probably lead to problems for individual households, but would hardly cause sufficient loan losses to threaten the banks' stability.

However, there is every reason to be vigilant with regard to continued developments, particularly from a price stability perspective. Continued rising asset prices and credit expansion could risk leading to overheating and thus rising inflation. An increase in the repo rate would in this case contribute both to counteracting incipient inflation tendencies and to reducing the risk of financial imbalances, through a dampening effect on demand for credit and on asset prices.

Footnotes

1)Excluding repos
2)For example, a 5-year goverment bond