Financial Stability 1/2004

Print

2 Macroeconomic developments, households and enterprises

2.1 Developments in the Norwegian economy
2.2 Households
2.3 Enterprises

Boxes:
- Fixed-interest mortgages
- What drives house prices?
- Predictions with two credit risk models

2.1 Developments in the Norwegian economy

Activity in the Norwegian economy is picking up. In Inflation Report 1/2004, it was projected that mainland GDP growth would be relatively high in 2004 and over the next two years compared with 2003 (see Table 2.1). The international recovery has also gained a firmer foothold.

Table 2.1

Norges Bank’s key rate has been reduced by 0.75 percentage point since Financial Stability 2/2003 and by a total of 5.25 percentage points since December 2002. The key rate is now 1.75%, which is historically low. The krone depreciated through 2003 and at the beginning of 2004, but has appreciated in recent months (see Chart 1.9). The underlying rise in prices in Norway is very low.

Private consumption expanded by a good 3.5% in both 2002 and 2003. The growth rate at the end of 2003 and the beginning of this year points to stronger growth in 2004. Low inflation and reduced interest expenses are expected to contribute to high growth in household real disposable income in the period ahead. This, combined with the prospect of some improvement in labour market conditions, has induced households to remain optimistic about the future (see Chart 2.1).

Chart 2.1

Gross capital formation for mainland Norway fell by 4.9% in 2003. However, the growth rate picked up and was positive at the end of last year. Service industries accounted for most of the increase, but housing and manufacturing investment also rose. According to Statistics Norway’s business tendency survey, Norwegian industrial leaders have become increasingly optimistic in their assessment of the short-term outlook (see Chart 2.2).

Chart 2.2

Petroleum investment exhibited brisk growth in 2003, primarily fuelled by investment in onshore installations. The average price of oil was USD 28.9 per barrel in 2003, which is the highest average annual price in 20 years. In April 2004, the average price was USD 33.4.

Since June 2003, the number of unemployed has gradually declined. In February 2004, seasonally adjusted LFS (Labour Force Survey) unemployment stood at 4.3%, compared with 4.6% on average in the second half of 2003. Unemployment is projected to edge down over the next few years.

Credit to mainland Norway as a percentage of GDP is at a historically high level (see Chart 2.3). Domestic credit to households is now increasing by around 11% at an annual rate (see Chart 3 in the Summary). The growth rate has ranged between 9% and 12% in the past four years. In February 2004, total credit to enterprises showed a year-on-year decline of 2.5%. The decline primarily reflects a fall in credit from foreign sources (see Chart 2.4).

Chart 2.3     Chart 2.4

There are a number of uncertainties associated with developments in the Norwegian economy in the period ahead. We have little experience of such substantial and rapid monetary policy easing as that seen over the past year and a half. A rising household debt burden means that private consumption will be more sensitive to interest rate changes. In addition, the interest rate level and the krone exchange rate influence enterprises’ competitiveness and earnings.

2.2 Households

Continued sharp growth in both debt and financial assets

The increase in household debt is largely matched by an increase in financial assets (see Table 2.2 and Chart 2.5). Household debt rose by NOK 110bn from the fourth quarter of 2002 to the fourth quarter of 2003, while gross financial assets increased by NOK 154bn.

Table 2.2     Chart 2.5

In order to analyse household transactions, we can disregard changes in insurance claims in group insurance schemes, which to a large extent consist of contributions by enterprises, and valuation changes as a result of changes in securities prices. Excluding such factors, household financial assets rose by NOK 80bn in 2003 (see Chart 2.6). Net investments in financial assets excluding group insurance claims have thus fallen to a negative NOK 30bn on an annual basis, from a negative NOK 5bn at the end of the 1990s. This implies that borrowing is now being channelled to consumption, investment in dwellings and other investments in fixed assets to a greater extent than to financial investments.

Chart 2.6

There is considerable uncertainty associated with calculations of households’ net investments in financial assets. They amounted to NOK 3bn in 2003 according to Norges Bank’s financial accounts, while according to the national accounts they were about NOK 31bn. The low figure in the financial accounts may indicate that disposable income has been lower or that consumption and/or households’ gross investments in fixed assets have been higher than assumed in the national accounts.² The assessment of developments in household income and saving in recent years has been particularly difficult due to substantial fluctuations in share dividends as a result of the abolition of the tax on dividends followed by uncertainty concerning a reintroduction of dividend taxation. According to the national accounts, the household saving ratio rose from 4.2% in 2001 to 9.4% in 2002 before falling again to 7.5% in 2003. This must be viewed in connection with the increase in dividend payments to households, from NOK 13bn in 2001 to NOK 43bn and NOK 35bn in 2002 and 2003, respectively (preliminary figures). Total saving came to NOK 28bn, NOK 69bn and NOK 57bn in the three years, respectively.

House prices on the rise

Higher house prices are an important explanation for the sharp rise in household debt (see Chart 2.7).³ Mortgage loans account for more than 70% of household borrowing. The share of households that own their own dwelling is high in Norway compared with other countries. After falling slightly for the first half of 2003, house prices have risen at an annual rate of 10% in the last six months (see Chart 4 in the Summary). Since autumn 2003, the time it takes to sell a dwelling has also been reduced compared with the same period one year earlier. The driving forces for developments in house prices are discussed in a separate box on pages 22-23.

Chart 2.7

House prices are high in a historical context (see Chart 2.8). Deflated by the building cost index, house prices are 16% higher than in 1987. Deflated by the annual wage index, house prices are marginally lower than in 1987. However, viewed in relation to disposable income, which in addition to wages is influenced by employment and other income, house prices are still somewhat lower than in 1987.

Chart 2.8

Increase in housing loans noticeably higher than housing investment

Households can raise loans with collateral in dwellings that have increased in value. The loans may also be used for purposes other than house purchases and home improvements, such as consumption, repayment of debt, financial investments and other investments in fixed assets. When the net change in the stock of housing mortgages exceeds housing investment, mortgage equity withdrawal occurs. This has been widespread in, for example, the UK, the US, Australia, the Netherlands and Denmark. The increase in the value of dwellings, combined with low interest rates, has also resulted in mortgage equity withdrawal in Norway in recent years (see Chart 2.9). Mortgage equity withdrawal in Norway probably takes place largely in connection with house purchases and may therefore increase with the number of house sales. The housing loan survey conducted by Kredittilsynet (the Financial Supervisory Authority of Norway) shows a rising proportion of loans for refinancing and purposes other than house purchases, particularly in 2003 following the substantial decline in interest rates. A sample survey conducted by Kredittilsynet also indicates that growth in consumer loans from finance companies slowed in 2002 and 2003. The increase in the value of dwellings has contributed to better borrowing terms for refinancing home mortgages than for new consumer loans.

Chart 2.9

Rising debt burden, but falling interest burden

Households’ debt-servicing capacity depends on their income, the size of debt and the interest rate level. The household debt burden, i.e. loan debt in relation to disposable income, rose rapidly in 2003 and is beginning to approach the high levels recorded at the end of the 1980s (see Chart 2.10).

Chart 2.10

Only the liquid portion of disposable income can be used to service debt. The debt burden figures will be higher if the return on insurance claims, which are generally illiquid, is deducted from income. The return on insurance claims has fallen as a share of disposable income since the 1980s when interest rates were high. The difference between the debt burden and adjusted debt burden was therefore greatest in the 1980s.

The household interest burden, i.e. interest expenses after tax in relation to liquid disposable income plus interest expenses, has declined in spite of the high rate of debt accumulation (see Chart 2.11). The main reason is the substantial decline in interest rates. The interest rate burden was reduced by 2.5 percentage points from the fourth quarter of 2002 to the first quarter of 2004.

Chart 2.11

Debt burden in different groups of households

Developments in the interest burden of various groups of households have varied. It is particularly low- and middle-income households (deciles 1-6) that have seen an increase in their debt burden compared with the 1980s (see Chart 2.12). Households with higher income (deciles 7-9) have the highest debt burden. The debt burden in decile 10 was considerably higher than for other groups in the 1980s, but has since fallen markedly. One of the reasons is probably changes in the tax system at the end of the 1980s (reduction in tax deductibility of interest expenses, particularly for high-income groups) and in 1992.⁴

Chart 2.12

The debt burden rose in all income groups in 2002 (latest available data), with the exception of decile 10. High dividend payments contributed to higher income in 2002, particularly for households in the top income group, who own most shares. This pushed down the debt burden for decile 10 in 2002. From 2003, the debt burden has been projected using growth in the average debt burden based on macro figures (see Chart 2.10). Dividend payments were probably high again in 2003. This may contribute to slower growth in the debt burden in 2003 than in the projections, particularly for groups with the highest income.

The distribution of debt by age shows that it is in the higher age groups (over the age of 55) that the debt burden has increased since the period prior to the banking crisis. The debt burden is nevertheless still lower in these groups than in the other groups (see Chart 2.13). The rise in the debt burden in the age group 25-34 must be seen in connection with the start-up phase they are in and the sharp rise in house prices.

Chart 2.13

It may have become more common and accepted to leave mortgaged assets to the next generation. Increased debt in higher age groups may also be based on a desire to adapt current consumption to expectations concerning future income. The financial buffers, financial assets in relation to debt, also showed an increase at the end of the 1990s, particularly in the age group 55-66. The financial buffers dropped from 2001 to 2002 for all age groups.

Developments ahead

Growth in household debt has been high for a long period. The projections are based on the assumption of a gradual deceleration of household debt growth to the same level as growth in disposable income at end-2006. The assumptions underlying the baseline scenario in Inflation Report 1/2004 have otherwise been applied, including the assumption that unemployment edges down. In 2006, the household debt burden approaches the level prevailing at the end of the 1980s (during the banking crisis). Partly as a result of low interest rates, the household interest burden will remain fairly low in spite of high debt growth.

If household debt continues to expand at the current rate, the debt burden will increase rapidly and in the course of 2005 will exceed the level during the banking crisis (see Chart 2.10). The interest burden rises more than in the baseline scenario, but is relatively low due to low interest rates.

If the household interest burden were to reach the same level as at the end of the 1980s, interest rates would have to increase by 7 percentage points in relation to the baseline scenario at the end of 2005 (see Chart 2.11). This is 8 percentage points higher than today. As a simplification, it is assumed that other variables that influence the interest burden are the same as in the baseline scenario. In the calculations, household interest income rises, but inasmuch as household debt is greater than their interest-bearing assets, interest expenses rise at a faster pace.

All in all, the decline in interest rates has made it easier for households to service their debt. At the same time, growth in household debt remains strong and far higher than income growth. Household financial assets are also rising. However, debt and assets are unevenly distributed among households. Asset values often fall during downturns, while the value of debt remains the same. High and rising debt therefore makes households more vulnerable to economic disturbances such as higher interest rates, an increase in unemployment and a fall in house prices. However, flexible inflation targeting in monetary policy reduces the possibility that households are exposed to a dual shock in the form of higher unemployment and higher interest rates, as was the case at the beginning of the 1990s.

Box:

Fixed-interest mortgages

With fixed-interest loans, the borrowing rate is set for a specific period. In periods when interest rates are generally rising, fixed-interest loans may contribute to financial stability because servicing the loans does not (immediately) become more demanding. On the other hand, borrowers with fixed-interest loans will not benefit from a fall in interest rates.

Historically, fixed-interest mortgages have not been common among Norwegian households. In 1994, fixed-interest mortgages accounted for 9% of the stock of loans to the household sector. A survey conducted by the Savings Banks’ Association in 2003 estimated that fixed-interest mortgages accounted for 20% of their stock of housing loans to households. The share is low compared with other countries (see Chart 1). In Norway, the trend over time is the opposite of what is observed in Sweden and Denmark where shares of fixed-interest mortgages have fallen, but from considerably higher levels.

Chart 1

The share of fixed-interest mortgages among new housing loans varies considerably from year to year (see Chart 2). The term of fixed-interest mortgages in Norway is often between 3 and 5 years.

Chart 2

A fixed-interest mortgage removes the uncertainty associated with interest expenses during the term of the loan and contributes to the predictability of the borrower’s expenses (“security motive”). Both the level of the fixed-interest rate and the fixed-interest rate relative to the floating interest rate may have a bearing on the choice of period for which the interest rate is fixed (“speculation motive”). Although long-term interest rates are historically low now, the difference between long-term interest rates and floating interest rates is considerable, so that these two factors have opposite effects.

The fairly limited number of fixed-interest mortgages in Norway may be due to a number of factors connected with both supply and demand:

• Borrowers may want to have the option of changing their repayment plan and the period for which the interest rate is fixed. In Norway, it can be costly for borrowers to terminate a fixed-interest mortgage agreement. By comparison, the option of earlier loan repayment without a penalty charge is a standard term of fixed-interest mortgage agreements in the US and Denmark.¹
• Private home financing in Norway has traditionally been offered by banks and not mortgage companies. A larger presence of mortgage companies would probably result in a larger number of fixed-interest mortgages, because these companies obtain their funding in the bond market.
• Fixed-interest mortgages increase banks’ interest rate risk if the banks have short-term financing. Small banks in particular may have lacked the expertise or financial resources to manage this interest rate risk.