Nyberg: The Baltic region in the shadow of the financial crisis
The global financial crisis has now been with us for over two years. It began in the summer of 2007 with anxiety about all the complicated credit instruments that had been issued using subprime US mortgages as collateral. However, the most acute phase of the crisis began almost exactly a year ago when the investment bank Lehman Brothers filed for bankruptcy protection. This was followed by such a rapid and extensive decline in liquidity in the international banking system that it led to a global financial crisis that also hit Sweden. The financial crisis accelerated and reinforced the economic downturn that had already begun. As a result, production around the world has fallen to an extent that we have not seen since the Great Depression of the 1930s.
The countries in the Baltic region, which are the subject of today's discussion, have also suffered. The Baltic states, that is Estonia, Latvia and Lithuania, have been hit hardest by the crisis. As a couple of the Swedish banks dominate the Baltic banking system, this has had consequences here in Sweden too.
Today, I intend to speak about how we at the Riksbank view the economic development of, and financial stability in, the Baltic region. The Baltic states and Sweden are naturally closes to our interests, so I will focus primarily on these countries. And as all financial crises are in various ways linked to the banks' loans on the property markets, I also see this seminar as an excellent opportunity to discuss these markets too.
We have heard and seen little more than negative news over the last 12 months and most forecasters, including the Riksbank, have underestimated the strength of the international economic downturn. Now, however, we are beginning to see increasingly distinct signs of a stabilisation of the global economy, above all on the financial markets. I therefore intend to conclude my speech by saying a few words about this.
The economic integration of the Baltic region is increasing
The Baltic region consists of eight countries. They are all members of the European Union (EU) and the region has a total population of almost 72 million. (1) This is more than 15 per cent of the total population of the EU member states. (2) Throughout the course of history, the countries in the region have had close links with each other, not least because of common economic interests. The Hanseatic League is a thing of the past, but the ties between the countries have not weakened, quite the reverse. The economic links are now stronger than ever and a large part of the countries' trade takes place within the region.
However, the integration does not relate to trade alone. Over the last ten years, Nordic banks have increasingly established operations in other countries in the region and have begun to offer their services to households and companies across national borders. Swedbank, for example, accounts for almost 45 per cent of total lending in Estonia, while SEB accounts for 30 per cent of lending in Lithuania. (3) Prior to the beginning of the crisis, the operations in the Baltic states also generated an increasing proportion of the earnings of these two banks (see Figure 1 and Table 1). Other major players are Nordea and Danske bank, which can now regard practically the entire Baltic region as their domestic market.
This financial integration has many economic advantages, but it also poses major challenges. I do not intend to discuss this further here, but will simply say that when the financial systems of different countries become intertwined, the risk increases that problems will spread across national borders. As far as Sweden is concerned, this has resulted in a decline in confidence not just in the Baltic economies but in the Swedish economy too. This is because there is a risk that developments in the Baltic states will entail substantial loan losses for the Swedish banks. This is one of the reasons why the Swedish krona has weakened against the euro. In the period since the beginning of the crisis to the present date, the krona has depreciated by approximately eight per cent against the euro. However, the fall has been even sharper over the last 12 months, reaching 18 to 19 per cent in March 2009.
Rapid downturn in the Baltic states
It became increasingly clear that the rate of growth in the world was beginning to slow down already in 2008. But it was not until the autumn, when the global financial crisis really took hold, that the situation seriously deteriorated. At present, the global economy is in recession and GDP is expected to fall by one per cent this year. World trade is expected to fall by over 12 per cent. (4) GDP is expected to fall in our region too – by 5 per cent in Sweden, which is slightly more than the figure for our Nordic neighbours. Development in Germany is somewhat weaker than in Sweden, while zero growth is expected in Poland. The situation is much worse in the Baltic states (see Figures 2 and 3), where GDP is currently expected to fall between 15 and 20 per cent this year. (5) This means that the crisis in the Baltic states is unparalleled compared with previous crises. The labour market has also weakened dramatically throughout the region. This is evident not least in the Baltic states where unemployment has now reached almost 17 per cent. Only a year ago, the level of unemployment was approximately 5 per cent (see Figure 4).
For several years, development in the Baltic states could be described as a fantastic success story with high economic growth. In the period 2005 to 2007, for example, economic growth in Estonia and Latvia was approximately 10 per cent per year, while development in Lithuania was more moderate. At the same time, however, major domestic imbalances developed which laid the foundations for the crisis we are witnessing today.
When the Baltic states moved from being regulated economies to market economies, optimism about the future was high and both household consumption and private sector investment increased dramatically. However, the investments went to the service and property sectors instead of to developing production capacity in the industrial sector. Export development was thus weak while imports increased rapidly as a result of strong domestic demand. Consequently, the current account deficits of the Baltic states increased (see Figure 4). The deficits were largely funded by taking loans in euro, mainly from Swedish and other foreign banks. The high level of economic growth led to the overheating of the labour market and wages and inflation increased (see Figure 6). In Latvia, for example, nominal wages in the public sector increased by more than 35 per cent in the year prior to the outbreak of the crisis, and in the private sector by over 30 per cent (see Figure 7). As, however, nominal interest rates remained low, the high level of inflation entailed negative real interest rates, which further fuelled the expansion of credit. Even though credit began to expand from a very low level, the rate of development was startling and it did not take long before the debts of the companies in relation to GDP reached levels only slightly below the level of private sector indebtedness in Sweden (see Figure 8). Household indebtedness did not increase quite as much (see Figure 9). As all three countries had also in various ways tied their currencies to the euro with the aim of eventually joining the European Monetary Union, they were not able to use monetary policy to restrict demand. The interest rate weapon was limited to the aim of maintaining the fixed exchange rate. At the same time, fiscal policy was expansionary and accelerated development.
Unlike the other countries in the Baltic region, the Baltic states thus entered the global financial crisis with several underlying weaknesses. There were large savings deficits in both the private and public sectors and a high level of indebtedness in the private sector. The loans also gave rise to a growing currency crisis. In addition, increasing cost pressures gradually undermined the competitiveness of the three countries.
When the global financial crisis broke out, economic growth in the Baltic states had already begun to decline. Declining growth and the large current account deficits led investors to begin reappraising the value of eastern European and central European assets. Capital inflows dried up, which made it more difficult to fund many projects and further reinforced the downturn in economic activity. The credit rating agencies downgraded the Baltic states, one after the other, which further aggrevated the shortage of capital. The lack of capital thus helped to reinforce the downturn in the same way that the surplus of capital once reinforced the dramatic upturn.
There are some signs that the real economy in this part of the Baltic region has now begun to stabilise, but unfortunately the recovery may be a long, slow process. This is partly because external demand is still weak as a result of the global recession. Above all, however, it is because domestic demand is low due to the fact that the Baltic states are now carrying out so-called internal devaluations. This means that wages and other costs have been substantially reduced to restore competiveness and to counteract the serious weakening of public finances. In the case of Latvia, this is taking place within the framework of a financial support programme led by the International Monetary Fund (IMF) and the EU. The previous imbalances have now declined in all of the three countries. The current account is, for example, now showing a surplus following years of deficits in all the three countries. This is mainly because there has been a very substantial fall in imports. Inflation has also declined, and in Estonia prices are now falling on an annual basis. However, these developments have also taken place at the expense of increased unemployment and impaired credit quality. This is reflected not least in the fact that late payments have increased dramatically.
Although there are signs that the downturn is slowing down, the development of the Baltic states in the period ahead is still associated with several risks. The main challenge relates to government finances where the budget deficits have grown as economic activity has weakened. The possibility to issue government securities to cover these deficits is, however, limited and Latvia is almost entirely dependent on the international support programme. Here there is also a great need for reforms in the public sector, reforms that should have been carried out during the period of positive growth. Estonia is in a better initial position as during the good years of the decade the country took the chance to build up fiscal policy reserves that can now be used as an extra buffer to fund the budget deficit.
Falling prices on the Baltic property market
In terms of financial stability, the development of the Baltic property market is of course a cause for concern. This development is in many ways reminiscent of the Swedish property crisis. Exaggerated expectations of future price increases pushed up prices, particularly housing prices. Now the prices of both commercial properties and housing have fallen dramatically. For example, the average price of an apartment in Tallinn was more than 50 lower in June 2009 compared to the peak in April 2007. This means that prices will soon be back at the level of 2004 (see Figure 10). In Latvia, the average price of apartments fell in the same period by almost 60 per cent, which is among the largest price falls noted in Europe during this crisis. (6) In Lithuania, the fall in prices has been comparatively moderate: 28 per cent during the second quarter of this year compared to 2008. (7)
We could say that after being a landlord's market for several years, the property market is now a tenant's market. In an economic downturn, it is common for landlords to offer rent rebates or to refrain from increasing rents in line with the index to the full extent allowed in rental contracts. Lenders can also put pressure on the construction and property companies to increase the level of occupancy. This can lead to lower rents than those the construction cost estimates were based on and thus give rise to payment difficulties for construction companies and property companies.
As I have said, the demand for properties in the Baltic states is very low at present. Eventually, however, the countries will recover and demand will return. There is also a strong underlying demand for property in these countries as much of the property there is old (it was built in the Soviet era) and in need of renewal. As property prices are already so depressed, this may mean that prices will increase rapidly when the economic turnaround finally comes.
Property also has a lasting value, providing that it is maintained. To the extent that the banks in Estonia, Latvia and Lithuania have taken properties as collateral it is probable that these properties will not be sold until there is an upswing in demand. This is what happened following the Swedish property crisis in the early 1990s when Securum and other companies were specifically set up to manage the settlement of bad loans with property as collateral. The increasing market value of property in the Baltic states will also eventually lead to an increase in the supply of new properties. An important precondition for all this is, however, that the internal reform programmes of the Baltic states are successful so that costs do not begin to rise once more.
Small risk of a crisis on the Swedish property market
In recent years, we have once again seen rapid price increases for commercial properties in Sweden. These prices have now begun to fall, however, and the question is whether this can lead to major loan losses for the banks in the same way as it did during the crisis of the 1990s (see Figure 11). The current downturn is after all much worse than that of the 1990s. I do not believe, however, that the effects on the property market will be so serious.
The years prior to the property crisis of the 1990s were marked by a rapid rise in the price of office premises. This was mainly a result of the deregulation of the credit market as a pent-up demand for loans led a rapid expansion of credit. A large part of the loans went to the property sector, where optimism was high as a result of the favourable situation on the market for office premises. The good times on the property market signalled the start of a construction boom. However, when the downturn came, prices fell and real interest rates increased the crisis was upon us. As well all know, this entailed substantial loan losses for the banks.
Today too, the property companies account for the banks' largest single exposure – almost 20 percent of lending goes to this group of borrowers. This is, however, a lower figure than during the crisis of the 1990s when the banks' exposure to the property companies amounted to one third. But, despite the fact that the percentage is now lower, exposure is still at such a level that it can constitute a potential credit risk for Swedish banks. A similarity between the increase in prices we have just seen and the increase at the end of the 1980s is that prices were not driven by so-called fundamental factors. In other words, prices increased not because of a higher demand for offices, lower vacancy rates or higher rents but rather because the investors did not demand as high a return on their invested capital. This in turn indicates that higher rents were expected in the future. (8) The risk premium was, for example, close to zero at the end of 2007. Investors did not thus compensate for the extra risk that a property investment entails compared to a risk-free investment such as a government bond. There were, however, periods in the 1980s when the risk premium was negative. The investors accepted a lower return on high-risk property investments than on safe investments in government bonds (see Figure 12).
However, there are also major differences between the situation on the Swedish property market then and now. One of the important factors that made it possible for prices to increase so much during the latter part of the 1980s was the deregulation of the credit market. Although lending to property companies has increased during the 2000s, this has not been to the same extent as in the years prior to the property crisis. The construction of new properties has also been moderate in recent years The most important difference, however, is that the interest rate is much lower today now that monetary policy no longer needs to focus on defending a fixed exchange rate (see Figure 13). The banks' credit assessments also focus more on the customers' cash flows and their ability to service their loans. In the 1990s, the banks focused to a greater extent on the value of the collateral, and as the collateral for the loans was often in the form of property the value of the collateral fell quickly in pace with the fall in prices.
With these differences in mind, we can note that it is unlikely that the fall in prices we have seen so far will lead to a crisis on the Swedish property market.
However, an economic downturn is always associated with increased risks. During the second quarter of this year, nominal prices for office premises in Stockholm, Gothenburg and Malmö fell by between 14 and 20 per cent on an annual basis. These are substantial price falls, although not as substantial as during the crisis of the 1990s. Prices began to fall in late 2007 and this was primarily driven by that fact that investors increased their return requirements. At the same time, rents were still increasing at a good rate, which counteracted an even more substantial fall in prices. However, rents began falling in the spring, especially in Stockholm where the number of vacancies has also increased. Given that the outlook on the labour market is deteriorating and that bankruptcies are expected to increase among tenants, it is highly probable that vacancies will continue to increase. Such a development will reduce the earnings of the property companies, thus weakening their ability to make interest and amortisation payments. Ultimately, this may also lead to the suspension of payments on the part of property companies too. Falling prices may also cause difficulties for the property companies if they do not meet the banks' requirements regarding maximum leverage. Some property companies, which before the current crisis funded parts of their operations abroad, appear to be experiencing difficulties with their funding.
Let me now say a few words about the Swedish housing market. Here, prices have not fallen in anything like the same way as for commercial properties. Despite the fact that we are experiencing the severest slump of the post-war period, house prices fell by only approximately two percent during the second quarter of this year compared to the same period last year. House prices in Sweden have so far fallen much less than in a number of other industrial countries (see Figure 14). Several of these countries, for example Ireland, Spain and the UK, have, however, had a much more ioverheated housing market than Sweden. Another reason why house prices in Sweden have not fallen as much may be that the Swedish banks have not been hit so hard by the financial crisis that they have found it difficult to lend money to the households. Incomes have also developed positively - at least for those who still have jobs. Another important explanation of why house prices have not fallen more is that the real interest rates faced by the households are at historically low levels. The construction of new housing has also been at a low level since the crisis of the 1990s. This is despite the fact that in most of the large urban areas it pays to build new house in comparison to what it costs to buy an existing house. (9)
At the moment, when the crisis appears to have bottomed out and optimism has begun to grow, housing prices are beginning to increase again, especially in the major cities. The Riksbank will closely monitor developments in the period ahead with regard to both the housing market and the market for commercial properties.
The Swedish banks can cope with large losses in the Baltic states
In the deep recession that we are now experiencing, it is not possible to dismiss the fact that the Swedish banks are facing significant loan losses. During the first six months of this year, these losses amounted to SEK 30 billion, which translated into annual figures is equivalent to 0.84 per cent of total lending. A large part of these losses, 44 per cent, stem from the Swedish banks' operations in the Baltic states. The Riksbank's assessment in the main scenario of the latest Financial Stability Report is that loan losses are expected to amount to just over 2 per cent of the banks' lending in 2009 and 2010. In a stress scenario that assumes a much worse situation in Sweden and in the Baltic states, the loan losses could amount to 4.3 per cent of lending during these two years. Almost 40 per cent of these losses in both the main scenario and the stress scenario can also be related to the banks' lending to the Baltic states and the rest of Eastern Europe. These are significant losses, even though they are not as large as those suffered during the Swedish bank crisis in the period 1991-1994 when losses amounted to 13 per cent of lending.
The banks' losses will certainly be largely related, in this crisis as in all others, to the situation on the property market. This is because most of the banks' collateral volume consists of property. However, the problems usually relate primarily to commercial properties, not to housing. The households service their interest and amortisation payments in most cases, even though they may sometimes have to negotiate a respite with the bank. Neither in Sweden nor in the Baltic states can households, as they can in the USA, give up the property they have provided as collateral and thus get rid of their debts. Not even during the crisis of the 1990s did the households account for a significant part of the banks' losses. However, although loans for housing seldom constitute major problems for the banks, they may mean disaster for individual households that suddenly find it difficult to make the payments. At present the interest rate is very low. However, our forecast indicates that the Riksbank, following a period with a low interest rate, will need to raise the interest rate quite quickly in the period ahead. It is therefore particularly important that the banks check that every household that borrows money can cope with the expected interest rate increases and potential losses of income.
The major Swedish banks have strengthened their capital bases and in international terms appear to be well capitalised. Three of the four major banks have carried out new issues since last autumn and Swedbank will soon conduct an additional new issue – all on the private market and without government guarantees. In many countries, where the banking systems have been affected by the current crisis to a greater extent, this has been impossible. In addition, all four major banks have strengthened their capital by reducing or cancelling dividends to shareholders for the previous year. It should be pointed out, however, that the danger has not yet passed. If the situation deteriorates more than expected, credit ratings may come under pressure and the banks may experience funding difficulties once again.
The Riksbank's measures safeguard financial stability
Since the financial crisis broke out, the Riksbank has entered into swap agreements with two of the Baltic states. We have made it possible for the Estonian central bank to swap Estonian kronor for Swedish kronor should the need arise, and for the Latvian central bank to borrow euros direct from the Riksbank. The aim of these agreements is to make it easier for the central banks in Estonia and Latvia to safeguard financial stability in the respective countries. Entering into these swap agreements also serves to safeguard financial stability in Sweden.
In order to secure the supply of liquidity, the Riksbank has also lent kronor and dollars to the Swedish banks when they have not been able to acquire funding on the market. These loans have been provided at longer maturities than normal and with a broader spectrum of collateral.
In Europe and other parts of the world, central banks and supervisory authorities have faced many new problems over the last 12 months, not least with regard to banks that have operations in several countries. The experience gained is now the subject of intensive discussion. I am convinced that in the future we will see major changes regarding the supervision and regulation of the banks and other financial operations. But I will gladly speak about this on another occasion.
Concluding remarks
Over the last 12 months we have experienced a financial crisis that is exceptional in many respects in the sense that it has hit more players, markets and types of asset than any previous crisis.
Since the beginning of the summer, the real economy has begun to show signs of stabilisation, although from very low levels. In Asia, and not least in China and India, growth has begun to pick up again and this is having a noticeable effect at those companies that export to these countries. World trade is beginning to stabilise and the fall in GDP has slowed down. The reports of the banks and companies for the second quarter have also been less negative than expected, which has given rise to somewhat greater optimism about the future. There are also signs that the financial markets are working more effectively. This is reflected, for example, in the fact that the credit spreads, that is the difference between the interest rates on corporate bonds and the interest rates on government bonds, have returned to the levels that applied before Lehman Brothers filed for bankruptcy protection. The difference between interbank rates and the expected policy rate has also decreased (see Figure 15). The credit markets are also steadily improving. Companies that just a few months ago found it difficult to issue bonds now have access to the market again. Companies with a good credit rating can now borrow at a lower cost, at longer maturities and with lower collateral requirements than before the summer. It thus appears that the situation on the financial markets is beginning to ease and that confidence is slowly returning. This is in turn a precondition for the continuation of the recovery of the real economy. Global monetary and fiscal policy is still highly expansionary and there is plenty of spare capacity in most countries. It is not unreasonable to assume that a rapid recovery of the financial markets can also facilitate a more rapid recovery in the real economy.
There are thus several reasons why it is reasonable to be cautiously optimistic. We must also remember, however, that we are still emerging from a severe recession and it may therefore take time before global demand recovers completely. Many banks in the world also have a lot of work to do before their balance sheets are free of impaired assets. From a Swedish perspective, economic development in the Baltic states still constitutes a significant risk.
Notes
1. I have chosen to define the Baltic region as Sweden, Denmark, Finland, Northern Germany (i.e. the federal states of Schleswig-Holstein, Mecklenburg-Vorpommern and Hamburg), Poland and the Baltic states – Lithuania, Latvia and Estonia.
2. Refers to the population of EU25.
3 Swedbank's market shares in the second quarter of thiss year amounted to 43 per cent in Estonia, 25 per cent in Latvia and 22 per cent in Lithuania. SEB's market shares in the same period amounted to 24 per cent in Estonia and 15 and 29 per cent in Latvia and Lithuania respectively. 4. IMF (2009) World Economic Outlook, June
5. Eastern European Consensus Forecast (2009), August
6. It is important to point out here that the statistics regarding property prices in the Baltic states are not complete and difficult to access, which may make if difficult to get a fair picture of the situation. This may mean, therefore, that both the dramatic increases and the substantial decreases that we are now seeing are exaggerated. It is also difficult to make direct comparisons between the countries, in particular over short periods of time, as property prices are measured differently in different countries. 7. Statistics on property prices are taken from the Bank of Estonia and Global Property Guide. 8.In simple terms one can say that the price of a property is determined by the expected operating surplus, that is rental income minus operating and maintenance costs, and by the investors' return requirement, which is the risk-free interest rate plus a risk premium. Investors are prepared to pay more for a property if they expect the rents to increase or if they do not require the capital they invest to generate such a high return.
9. In the economic literature this ratio is called Tobin's q after the Nobel Prize winner James Tobin who, among much else, has studied property prices. If Tobin's q is equal to 1, then the cost of buying an existing house is equal to the cost of building a new one. This means that if house prices increase so that the market value is higher than the replacement value, then it pays to build a new house as each invested krona is valued at more then one krona by the market. In several of our large cities, Tobin's q is between 1 and 2, or even higher. In many parts of Sweden, however, the value is well below 1, which means that it does not pay to build new housing there.
