Subprime loans in the United States were the origins of the deep-seated tensions in the international financial markets, which have not yet relinquished their grip after almost a year. The problems in the subprime market are an important reason behind the collapse in housing construction and the falling house prices, which have made a profound impression on the real economy in the United States. In some other countries, too, the earlier sharp rises in property prices have slowed down and been replaced by falling prices. Here in Sweden we have so far been spared these events. How does the Riksbank view future developments in the property market and what stance does the Bank's monetary policy take in relation to asset prices? Today I will address these questions. I will also briefly comment on the recent developments in the financial markets.
Several countries have experienced rapidly rising house prices, for instance the United States and the United Kingdom, as well as Denmark, Ireland and Spain. Recently there has been a slowdown in the rate of increase and in some of these countries house prices are now falling. In the United States house prices have fallen by around 10 per cent since the peak in 2006. (see figure 1)
In Sweden, too, house prices have risen rapidly, even more so than in the United States. Since 1995 Swedish house prices have risen by more than 150 per cent; the corresponding figure in the United States is around 115 per cent. At the same as house prices have soared, inflation in Sweden has been moderate and consumer prices have only risen by 14 per cent since 1995. (see figure 2)
There are several factors behind the upturn in Swedish house prices; Swedish households’ real incomes have grown, their financial wealth has increased and in addition real interest rates have been low. Although more housing has been built, it has not been enough to meet the demand. The percentage of Swedish municipalities reporting a shortage of housing doubled to 42 per cent between 2002 and 2007 according to statistics from the National Board of Housing, Building and Planning. More than half, or 60 per cent, of the population now live in municipalities with housing shortages.
Households’ debt ratio, that is their debts in relation to their disposable income, has risen rapidly in recent years and is now around 150 per cent. It is difficult to compare the current debt ratio with earlier periods, as these have been marked by the conditions prevailing at the time. For example, the debt ratio that preceded the credit boom at the end of the 1980s was marked by the regulated credit market and did not reflect either households' willingness or ability to bear debt. The period after this was affected by the economic crisis and institutional changes with an ensuing rapid credit crunch. Unlike the 1980s, the build up of debt in recent years has been in an environment of low inflation and low interest rates. The interest ratio, that is, interest expenditure as a share of disposable incomes, is much lower now than it was then.
Although the level of households' debt ratios is not necessarily worrying, it is nevertheless the case that a borrowing level rising by more than 10 per cent a year is not sustainable in the long term. In the long term it needs to come down and be more in line with nominal disposable incomes. The Riksbank is assuming that household borrowing will continue to increase, but at a slower rate than before. The fact that interest rates have been raised contributes to this, as does the fact that economic activity is entering a calmer phase. The interest ratio has recently begun to rise but is still at an historically low level. The expected slowdown in household borrowing will mean that the interest ratio also increases at a slower rate in future. (see figure 3)
The chances are good that household borrowing and house prices will have a soft landing. Although borrowing is still increasing at a rapid rate, there has nevertheless been some slackening and this is expected to continue. House prices fell slightly in the first quarter of this year compared with the final quarter of last year. However, on an annual basis the increase was just over seven per cent. Prices of owner-occupied apartments also fell during the autumn, but have since stabilised.
Employment and household incomes are of the utmost importance when it comes to movements in the property market. According to the Riksbank’s most recent forecast, employment is expected to continue rising this year, albeit at a slower pace than last year. Following a strong rise last year, around 4 per cent, households’ real disposable incomes are expected to increase by a good 2 per cent over the coming years. Given this outcome, there is little risk of a fall in house prices to the same extent as is happening in the United States, for example. In Sweden savings are also high to start with, unlike the United States where they have fallen to historically low levels.
There is always a risk that house prices, and also other asset prices such as shares, which have been pushed up by exaggerated optimism or speculation, could plummet. It would appear that Swedish price movements can be explained by fundamentals, and that the speculative element has happily been conspicuous by its absence.
As property often functions as collateral for loans, a large fall in prices would affect the banks. In addition, developments in the property market affect demand in the economy. Rising house prices mean that households can take on new loans to increase their consumption. Correspondingly, falling house prices can contribute to subduing demand in the economy. An IMF study of twenty property crises in different countries shows that they entailed a fall in prices of 30 per cent. The countries’ GDP was on average four per cent lower three years after a property crisis had started than it would have been without the fall in prices(1). The scars in the real economy may well be considerable and given this it is interesting to discuss how a central bank should react to changes in asset prices.
In the international debate on asset prices and monetary policy there are several different points of view. A common view is that monetary policy should only react to changes in asset prices to the extent that these affect the central bank’s inflation and growth forecasts. One should thus not have a separate strategy for dealing with changes in property prices.
Others think that a central bank can improve the outcome for inflation and production by reacting to changes in asset prices over and above their consequences for the inflation forecast. Two main arguments have been put forward. Firstly, monetary policy can alleviate the effect of the price change on the real economy. Secondly, the actual knowledge that the central bank will react should reduce the probability of exaggerated price movements occurring. The difficulty lies in determining whether or not prices are increasing at an exaggerated rate. It is not necessarily so that a central bank is better qualified to make such an assessment than the market participants. In addition, there is uncertainty over how monetary policy affects asset prices.
The Riksbank conducts what is known as flexible inflation targeting. The objective is that the annual change in the consumer price index (CPI) shall be 2 per cent, but we also attach some importance to attaining a balance in production and employment. Asset price movements are normally included within the scope of flexible inflation targeting. We include changes in asset prices and other financial variables in the balance when we make our forecasts. However, the Riksbank has no target for property prices.
The Riksbank’s view is close to the reasoning that changes in asset prices should only be allowed to affect monetary policy to the extent that they affect our forecasts. But such a view is based on the forecasting models used being able to include the full effect of house prices on total demand in the economy. This is not the case for the Riksbank or other central banks; there are no forecasting tools that can entirely capture the effects of changes in property prices. The forecasts therefore have to be supplemented with qualitative assessments.
How has the Riksbank’s stance affected monetary policy in practice? When the Riksbank raised the repo rate in January 2006 borrowing and house prices had a more pronounced effect on the decision. The Riksbank assumed that it would soon be necessary to raise the interest rate, but if only inflation had been taken into consideration, it would have been possible to wait a little longer before making this decision. Now the interest rate raise was brought forward to counteract the risk that an adjustment in the property market would cause a large fall in demand and lead to inflation undershooting the target when regarded in a perspective beyond the normal forecast horizon. The influence on interest rate policy was thus marginal, but on the other hand the expansion in household borrowing and house price movements were mentioned frequently in a number of speeches and interviews.
I was not at the Riksbank then, but I can imagine that it was no easy decision. The Riksbank was criticised for not bringing up inflation to the target of 2 per cent, that is, the repo rate should not have been raised. But could one just sit by and watch the risks that threatened in the longer term? This was a conflict of interests that soon petered out. The interest rate raises made since then have gone hand in hand with strong economic activity as well as rising inflationary pressures, which have demanded a higher interest rate.
Let me now also mention the commercial property market. Commercial property prices rose sharply during the second half of the 1990s and peaked in 2000. During the following four years prices fell. In 2005 they began to rise again and in Stockholm they have risen by 60 per cent up to the present time.
At the end of the 1990s the price increases appear to have been largely driven by an increase in demand for office premises, by vacancies being cut down and by rents increasing. The situation during the upswing in recent years, from 2005, has been different. Rents have not followed the price upswing to the same extent as before, and prices appear to have been largely driven by a decline in investors’ demands for return. Investors have been requiring less compensation for the risk they take when investing in property. This means that investors appear to have had high expectations of future income, that is, rents. (see figure 4)
As in the housing market, the price increase on office premises slowed down somewhat in 2007 and this trend has continued during the first quarter of 2008. The same tendency can be found in several European countries. The financial turmoil has pushed up the price of risk in general, and this has also applied to the property market. In the United Kingdom the required rate of return increased quickly and property prices have fallen.
There are some indications that Swedish investors in property have begun to raise the price of risk. So far this appears to mainly apply to property in poorer locations and smaller towns. In Stockholm city centre investors in the property market are not requiring any higher earnings than a risk-free five-year government bond provides. It may be the case that the required rate of return has begun to rise more in general, but that this has not been captured in the statistics. (see figure 5)
It is difficult to determine what direction prices for office premises will take. Rising risk premiums indicate that prices should be pushed down. At the same time, however, the rental market is currently relatively strong. As the service sector is office-intensive, activity in this sector affects the possibilities for leasing. But there is considerable sensitivity to weaker economic activity.
In the autumn Financial Stability Report the Riksbank saw a risk that expectations of future increases in income in the property market were too high. We thus saw the risk of a fall in prices. Since then the required rate of return has risen with regard to commercial property around the world, while economic activity is more uncertain. This should reasonably have an effect on prices in the Swedish property market. (see figure 6)
The property market remains a hot topic and those who are interested can read more about the property market and commercial property in the Riksbank's next Financial Stability Report, which will be published on 3 June.
Let me conclude with a few words on the financial markets. As I mentioned earlier, the problems in the US subprime market had widespread and profound effects on the international financial markets. These loans had been packaged together with other securities into structured products and sold to investors around the world. The uncertainty over where the potential losses were created a general distrust between the banks and a general increase in risk aversion.
The former was expressed in a rise in the differences between interbank rates and risk-free government bonds, what are known as TED spreads. The latter was expressed in, for instance, rising credit spreads. (see figure 7 and 8)
For a few weeks now we have seen signs that the tensions in the credit market have eased. The credit spreads have declined and the cost of insuring against losses has fallen. (see figure 9)
Other signs are that the banks have been able to sell off loans from their balance sheets. They have also been able to strengthen their own capital, although a large part of this process still remains. The stock markets have also recovered.
The uncertainty in the interbank market has prevailed longer; the TED spreads have remained high but have fallen over the past few days. In Sweden the TED spreads have also risen, but unlike other countries the markets have continued to function without any intervention from the Riksbank.
Despite the fact that there are positive signs, it is still too early to assume the danger is past. There may still be loan losses that may damage trust and confidence in the market when they come to light. This applies to both the United States and Europe.
Notes:
1) IMF World Economic Outlook, April 2003