Financial Stability 2/2002
3 Macroeconomic developments and credit risk
3.1 The macroeconomic environment
3.2 Credit growth
3.3 Credit risk associated with loans to the household sector
3.4 Credit risk associated with loans to the enterprise sector
Boxes:
- Commercial property market
- Market values and the risk of bankruptcy
3.1 The macroeconomic environment
Growth forecasts for the Norwegian economy presented in Inflation Report 3/2002
have been revised downwards somewhat compared to the previous report. Weaker
international developments, an increase in the key rate in July and an appreciation
of the krone are contributing factors. In a number of areas, the Norwegian economy
will be affected by the weak developments in the global economy, but factors
peculiar to Norway contribute to a mixed picture.
The outlook for the exposed sector is subdued. High wage growth combined with
a strong krone has resulted in a considerable deterioration in competitiveness
for these industries. High oil prices, which boost activity and investment,
contribute to sustaining output and demand in parts of the manufacturing industry.
In addition, sustained growth in activity in the service industries is expected
in the period ahead. Household real wage growth is high and households are still
optimistic about their own future financial developments (Chart 3.1). Fiscal
policy is also expected to fuel domestic demand in the period ahead.
Chart 3.1
In June 2002, Statistics Norway published the revised national accounts for
the period 1991-2001. As a result of the revision, the historic development
in some of the indicators in this report has been adjusted since last time.
Operating results for the enterprise sector have been revised upwards appreciably,
especially in the period 1996-2001. Consequently, the debt burden in the enterprise
sector is somewhat lower than calculated earlier. The changes in household income
figures have been small so that changes in the historic debt burden are insignificant.
The household saving rate is considerably lower than previously estimated because
consumption growth was higher in the latter half of the 1990s than assumed earlier.
3.2 Credit growth
Year-on-year growth in total credit to the public (municipalities, non-financial
enterprises and households) (C3) has been relatively stable at around 8% this
year. Credit from foreign sources has increased less than credit from domestic
sources since year-end 1999. Overall credit growth has been appreciably higher
than GDP growth in recent years. At the end of the second quarter 2002, overall
credit amounted to 140% of GDP, while overall credit to mainland Norway increased
to 163% of mainland GDP (Chart 1.1). Viewed in isolation, strong growth in credit
over a long period increases the risk of financial instability.
Chart 1.1
Year-on-year growth in domestic credit to non-financial enterprises has been
between 7% and 9% this year (Chart 3.2). This is still a substantial real growth,
but appreciable lower than in 2001. In September, year-on-year growth declined
to 6.6%. Growth in credit to the household sector in particular remains high.
Year-on-year growth in domestic credit to households has been over 10% since
March 2000 and was 11.2% at the end of September. Household borrowing in foreign
currency (from both domestic and foreign sources) has increased sharply the
last few years. However, foreign currency loans still account for a small portion
of overall household debt (about 1.5%).
Chart 3.2
3.3 Credit risk associated with loans to the household sector
Sustained high growth in household debt
Household debt continues to rise sharply. High income growth and considerable
optimism about the future increase household willingness to take on debt and
press house prices up. Real house prices are now 23% higher than the previous
peak level recorded in 1987. In relation to income growth, however, house prices
are still somewhat lower (Chart 3.3). Growth in household debt has remained
record high throughout 2001 and so far in 2002 even though the rise in house
prices has slowed in this period. At the end of October 2002, house prices were
about 7% higher than one year earlier. After seasonal adjustments, house prices
rose both in September and October. According to real estate industry statistics,
average turnover time was 32 days in September and October, down from 37 days
in August. Thus, available statistics provide few indications of a marked shift
in the housing market this autumn.
Chart 3.3
Figures from the Banking, Insurance and Securities Commission’s housing
loan survey show that the average loan-to-asset value ratio in connection with
house purchases rose in 2002. The share of housing loans (measured in terms
of value) with a loan-to-asset value ratio of more than 100%, i.e. loans that
are larger than the collateral value of the dwelling, has increased markedly
the last two years (Chart 3.4). At the same time, however, there has been an
increase in the share of loans that have been granted contingent on the provision
of additional collateral. Nevertheless, the value of total collateral (primary
plus additional) is lower than the loan amount on only four of ten loans with
a loan-to-asset value ratio over 100%.
Chart 3.4
Increased debt in vulnerable groups
Even though the overall debt burden (debt as a percentage of disposable income)
in the household sector is still lower than at the end of the 1980s, we showed
in Financial Stability 1/02 that the debt burden of the low- and middle-income
categories is higher than ever before. Payment problems for households in these
income categories may thus have a greater impact on banks than earlier.
Statistics Norway’s income and wealth survey for 2000 shows that an increasing
share of total household debt may be ascribed to households with a heavy interest
burden (interest expenses as a percentage of disposable income before deductions
of interest expenses). Many households have taken on loans which are large relative
to their disposable income. More than 30% of total household debt in 2000 may
be attributed to households with a relatively high interest burden (more than
20% of disposable income). The corresponding figure in 1997 was 17%.
As of 30 June 2002, total household debt was just over NOK 1000bn. If we assume
that the distribution of debt by income categories and the share of loans to
households with a high interest burden have remained unchanged from 2000 to
the first half of 2002, households with a relatively high interest burden will
account for about NOK 315bn in debt (Chart 3.5). Of this amount, approximately
NOK 108bn is debt in households with an income after tax of less than NOK 300
000 (deciles 1-6). In 1997, the corresponding risk-weighted debt in income deciles
1-6 amounted to about NOK 42bn.
Chart 3.5
Households in low- and middle-income categories with a high interest burden
may experience debt servicing problems in the event of a loss of income or an
increase in interest rates. As we showed in Financial Stability 1/02, households
in the low-income category have few financial assets on which to draw should
they experience payment problems. Thus, viewed in isolation, this development
implies increased credit risk for banks.
Reduced financial wealth
In the second quarter of 2002, household gross investment in financial assets
(calculated as the sum of the last four quarters) was equivalent to gross credit
growth in the same period (Chart 3.6). Thus, net investment in financial assets
was zero in this period. As a result of the drop in equity prices, the value
of household securities holdings fell by NOK 18bn from the first to the second
quarter of 2002, while the value of their insurance claims declined by NOK 2bn
in the same period. Overall net financial wealth declined by NOK 38bn from the
second quarter of 2001 to about NOK 365bn at the end of the second quarter 2002
(Chart 3.7). If wealth is adjusted for insurance claims, which are illiquid
and can only be used to a limited degree if households experience debt-servicing
problems, financial wealth is considerably negative (Chart 1.3)
Chart 3.6 Chart
3.7 Chart 1.3
Developments ahead
The household debt burden continues to rise as a result of sustained strong
credit growth (Chart 3.8). If the level of borrowing in relation to income growth
remains as high in the years ahead as in the last year, the average debt burden
at the end of 2004 will be as high as at the end of the 1980s. Despite the high
level of borrowing, the average interest burden has declined slightly the last
half year (Chart 3.9). The reason for this is somewhat lower interest rates
in the first half of the year.
Chart 3.8 Chart
3.9
Norsk Gallup’s consumer confidence indicator shows that household expectations
regarding their own future financial developments are still relatively high.
In relation to household expectations concerning own income growth, interest
rates may seem fairly low (Chart 1.2). If future income developments do not
match household expectations, accumulated debt will not “disappear”
as quickly as in the 1970s and 1980s, due to the low inflation rate. For this
reason, there is a greater need than earlier for households to think long-term
with regard to financial dispositions. Overall, credit risk associated with
loans to the household sector is considered to be moderate and somewhat higher
than in the last Financial Stability report.
Chart 1.2
3.4 Credit risk associated with loans to the enterprise
sector
Relatively moderate rise in debt, but continued rise in debt burden
Enterprise sector debt is now growing at a relatively moderate pace. For the
past half year, the year-on-year rise in enterprise debt has been stable at
about 7-9%. Borrowing trends of different sectors differ considerably (Chart
3.10). Bank lending to sheltered industries like retail trade and services has
increased more than lending to exposed industries. The debt situation in enterprises
is closely tied to their investment. Mainland business investment declined by
8.8% from the first half of 2001 to the first half of 2002. Investment tax was
abolished on 1 October this year and planned investments may have been postponed
in anticipation of this.
Chart 3.10
Growth in enterprise sector debt is still stronger than growth in value added.
The enterprise debt burden (interest-bearing debt as a percentage of the cash
surplus less interest expenses) has increased further (Chart 3.11). Their debt
burden is now as high as it was during the banking crisis. The enterprise interest
burden (interest expenses as a percentage of cash surplus) has also increased.
In 2001, interest expenses amounted to almost 50% of the surplus. The interest
burden is now at the same level as in the first half of the 1990s.
Chart 3.11
Profitability down in 2001…
Recently published corporate account figures for 2001 show that profitability
measured as enterprises’ return on total assets fell in most sectors,
but not all (Table 3.1). The fall was greatest in the IT, telecom and fish farming
sectors. Profitability improved in the mining, shipbuilding, utilities, construction
and shipping sectors. Return on equity fell appreciably in 2001 compared with
the previous year (Chart 3.12). The equity ratio dipped somewhat from 2000 to
2001, but remained relatively high.
Table 3.1 Chart
3.12
… and virtually unchanged probability of bankruptcy
Enterprise results and equity ratios fell in 2001 viewed as a whole. This is
largely due to developments in the enterprises that are least exposed to risk.
Our bankruptcy prediction model1 uses the revised
accounts figures for 2001 to show that the probability of bankruptcy for a median
enterprise in 2001 was roughly unchanged from the level in 2000 (Chart 3.14).
However, the probability of bankruptcy for the weakest enterprises was reduced.
The most vulnerable enterprises, i.e. those with a probability of bankruptcy
of more than 2%, improved their equity ratio in 2001 compared with the previous
year. The results of some of these enterprises also showed some improvement.
Chart 3.13
1 See Eklund, Trond, Kai Larsen and Eivind
Bernhardsen, “Model for analysing credit risk in the enterprise sector”,
Economic Bulletin 3/2001, Norges Bank.
Calculations by Moody’s KMV based on share prices and accounting data
show an increase in bankruptcy probabilities for the largest Norwegian non-listed
companies in 2002 (Chart 3.14). The increased risk of bankruptcy can be explained
by the decline in corporate assets and uncertainty regarding future earnings
(see separate box).
Chart 3.14
The share of bank debt in various risk groups changed little in 2001 compared
with the previous year (Chart 3.15). At end-2001, bank debt in enterprises with
an estimated bankruptcy risk of less than 1% accounted for almost 78% of total
bank debt.
Chart 3.15
Risk-weighted debt (debt multiplied by the probability of bankruptcy) is a
measure of banks’ expected loan losses in the absence of collateral. In
2001, risk-weighted debt increased in service sectors including IT, commercial
services and property management, and in production sectors including manufacturing
and fish farming (Chart 3.16). However, risk-weighted debt decreased in a number
of sectors, partly as a result of lower bankruptcy probabilities.
Chart 3.16
Risk-weighted debt in enterprises where the probability of bankruptcy was greater
than 2% edged up in 2001 to 78% of total risk-weighted debt. Most of this debt
is attributable to the retail trade, property management, manufacturing and
commercial services sectors. Geographically, most of this debt is attributable
to Oslo/Akerhus County and Western Norway. The negative developments in fish
farming have affected Western and Northern Norway. As a result of the negative
developments in the IT sector, debt associated with this sector now constitutes
a large share of debt with the highest risk level in Oslo/Akershus and Southern
Norway. The increase in risk-weighted debt in the property business has affected
Oslo/Akershus, Eastern Norway and Northern Norway in particular.
Commercial property market
The commercial property industry is the single largest recipient of bank
loans in the business sector. At the end of 2001, loans to property companies
accounted for NOK 135bn or 11% of total bank lending. Banks’ loans to
property companies are often secured against property belonging to the company.
Because of the close connection between the property company’s earnings
and the value of its property, changes in banks’ expected losses can
be rapid and substantial. In other words, developments in the property industry
have a considerable impact on the risk in banks’ loan portfolios..
Property companies are, however, a complex group, ranging from a group subsidiary
leasing property to the parent company to property companies that simply develop
and lease various types of premises. Important rental market segments are
offices, shops, restaurants, shopping centres and hotels. Lower demand for
commercial space from financially weak groups of tenants may in some cases
be replaced by increased demand from other groups experiencing growth in activity.
The decline in the equity market, for example, has had an impact on the financial
industry and parts of the consultancy sector. In addition, the outlook for
activity in the ICT industry is uncertain. However, expected growth in the
sheltered sector may boost demand for space.
Risk exposure for a property company depends on a number of factors in addition
to geographical diversification and the leasing of property to several types
of segments. Risk depends on the length of rental contracts and the spread
of rental due dates. In addition to the risk associated with rental income,
property companies’ profits are directly influenced by interest rate
changes and changes in property portfolio values.
A large share of banks’ losses during the banking crisis was linked
to commercial property. During the banking crisis years, rental prices for
commercial space were reduced and the equity ratio and debt servicing capacity
(earnings in relation to bank debt) of enterprises in the commercial property
market declined (Chart 1). In the years following the banking crisis, rental
prices increased and enterprises improved their debt servicing capacity and
equity ratio.
Chart 1
In recent years, the supply of commercial properties has risen. The figures
on commercial building starts indicate that this will continue. Property companies
have increased their debt, but have not increased their earnings/rental prices
to the same extent. Little information is available about price developments
in commercial property. This is partly because there are so many different
types of premises and price developments may vary from one to another. Figures
for commercial property in the Oslo area show that rental prices have fallen
over the past few quarters. Future price developments, and thus the debt servicing
capacity of property companies, will depend on both the supply of and the
demand for commercial property and thus on the general economic situation.
The equity market does not seem to be particularly worried. Although share
prices for listed Norwegian property companies fell in 2002, they fell less
than prices on the Oslo Stock Exchange as a whole (Chart 2).
Chart 2
Increase in defaults and bankruptcies
Gross non-performing loans in the enterprise sector, measured as a percentage
of the sector’s total bank loans, have increased this year. At the end
of the third quarter, there was a marked increase to 2.3% from 1.8% in the same
period last year. However, the level is still low compared with the first half
of the 1990s.
The number of bankruptcy proceedings initiated was 20% higher in the first
three quarters of this year than in the same period in 2001. The percentage
rise in market value of the enterprises that have gone bankrupt is even higher.
If fourth quarter developments are similar to developments so far this year,
the number of bankruptcies will exceed 4000. This would be a substantial increase
compared with recent years. Construction, property management, rental activities,
commercial services and fishing and fish-farming are among the sectors that
have had the largest increase in bankruptcies so far this year, compared with
the end of the third quarter of 2001.
Developments ahead
The sharp fall in equity prices in 2002 indicates that participants in securities
markets are expecting lower enterprise sector earnings. If enterprise earnings
decline and the interest rate increases, a number of enterprises may have debt
repayment problems. Enterprises in the exposed sector are most vulnerable, particularly
enterprises in sectors such as IT and manufacturing. Sheltered sector enterprises
will not be affected to the same extent. On average, enterprises’ equity
ratios are higher than at the beginning of the 1990s. If enterprise debt growth
continues, however, projections of enterprise accounts indicate that the probability
of bankruptcy will increase2 markedly for the
weakest enterprises from 2003 (Chart 3.13). The weakest enterprises, in particular,
will be affected by a prolonged period of low earnings. We expect an increase
in banks’ losses on loans to enterprises in the most exposed industries.
The credit risk associated with loans to enterprises is higher than it was half
a year ago, and must now be described as relatively high for loans to enterprises
in the exposed sector, but still moderate for enterprises in sheltered industries.
Chart 3.13
2 The projection method is described in Frøyland,
Espen and Kai Larsen, “How vulnerable are financial institutions to
macroeconomic changes?” Economic Bulletin 3/2002, Norges Bank.
Market values and the risk of bankruptcy
The risk of enterprise bankruptcy is a key factor in our analysis of the
credit risk associated with enterprises. Financial strength is an important
explanatory factor for this risk. Enterprises with sound financial strength
are capable of absorbing future losses. The equity ratio is a measure of the
financial strength of the enterprise, but there is a difference between equity
capital based on market values and equity capital based on recorded values.
Whereas market values reflect expectations with respect to future earnings,
accounting values are based on the completed accounting period and asset valuations
according to the historic cost method, i.e. they are retrospective.
Norges Bank’s bankruptcy prediction model SEBRA1)
is based on enterprise accounts and thus represents the latter approach. However,
models based on market values have also been developed.2)
These approaches complement one another. Particularly when economic conditions
change substantially, market values may reflect important new information
that is not yet reflected in the accounts.
1) See Eklund, Trond, Kai Larsen and
Eivind Bernhardsen: “Model for analysing credit risk in the enterprise
sector”, Economic Bulletin 3/01, Norges Bank.
2) This approach is also used in the
model developed by Moody’s KMV.
Calculating the equity ratio, i.e. the ratio of equity capital to the total
value of the enterprise, using the accounts, is not a problem. However, calculating
the equity ratio from market values is not as simple. The market value of
the equity capital of listed enterprises is known, but the market value of
the total assets of the enterprise (value of liabilities and equity capital)
is not directly observable. One approach is to use a pricing approximation
in which the total assets of the enterprise determine the value of the equity
capital. The total assets can then be derived from the observed value of the
equity capital. One such approach is to use option pricing theory. The owners
have a right to the enterprise’s assets after its debts have been paid.
This right can be regarded as an option on the enterprise’s assets with
a contract price equivalent to the enterprise's debt. Information from equity
markets can then be used to estimate the level and volatility of the enterprise’s
total assets.3)
3) See for example Merton, R.C., “On
the Pricing of Corporate Debt: The Risk Structure of Interest Rates”,
Journal of Finance, Vol 29, No 2, 1974, pp. 449-470 and Black, F. and M.
Scholes., ”The Pricing of Options and Corporate Liabilities”,
Journal of Political Economy, 1973, pp. 637-654.
This approach offers useful insight into the fundamental conditions that
influence the competitiveness of enterprises: how much debt the enterprise
has, expected developments in the total assets of the enterprise and uncertainty
associated with these developments. If the value of the assets is less than
the enterprise’s debt when it falls due for payment, the enterprise
will be bankrupt. For a given debt level, a decline in expected total assets
will increase the risk of bankruptcy, because it will be more probable that
the value of the enterprise is less than its debt (Chart 1). An increase in
uncertainty regarding future value has the same effect (Chart 2).
Chart 1 Chart
2
© 2002 Norges Bank webmaster@norges-bank.no