The Financial Sector

The financial sector in Denmark is still robust. Stress tests show that the resilience of banking institutions to higher losses remains sound due to record-high earnings and the increasing capital base.

Particularly among the small and medium-sized banking institutions the growth in lending continued to be high in 2006. The rapidly increasing deposit deficit, which is a result of the high lending growth, makes it necessary to focus on capital and liquidity management. The opportunity to use housing loans as collateral for issuance of covered bonds (SDO) may contribute a new, stable source of financing for the banking institutions.

The resilience of the mortgage-credit institutes to increasing losses has declined a little, but is still very high. Life insurance companies and pension funds have generally strengthened their reserves, even though investment yields were lower in 2006 than in 2005.

THE SIGNIFICANCE OF FINANCIAL INSTITUTIONS TO FINANCIAL STABILITY

The banking institutions play an important role in the financial sector as intermediaries of loans between depositors and borrowers. This chapter focuses on the banking institutions' earnings capacity, risks and resilience.

In Denmark there are two large, a number of medium-sized, and many small banking institutions. The two large banking institutions differ from the rest of the sector in terms of both size and business areas. Consequently, these two are compared with other large Nordic banking groups. The analyses in this chapter are based on 55 selected banking groups and institutions in three categories. Category A comprises the six largest Nordic banking groups, category B the 12 medium-sized Danish banking institutions, and category C comprises 37 small Danish banking institutions.^[1]

The mortgage-credit institutes provide credit to finance real estate, which makes them the largest bond issuers in Denmark. The development in the mortgage-credit institutes can affect the banking institutions' earnings through ownership and collaboration agreements, or via competition in the market for mortgage financing.

Similarly, life-insurance companies and pension funds (hereinafter pension companies) can influence the banking institutions via ownership and via competition in the pension market. At the same time, this sector plays an important role in the financial markets via its management of substantial funds.

NORDIC GROUPS AND DANISH BANKING INSTITUTIONS

Record-high earnings
The Nordic groups and the Danish banking institutions posted record-high results in 2006.

The Nordic groups in category A achieved a pre-tax profit of kr. 98.5 billion, cf. Table 1, equivalent to an increase of 14 per cent on 2005, and
primarily due to increased net fee income from asset management and securities activities, value adjustments, and reversal of write-downs on loans. Interest margins on loans continued to decline, but the impact on net interest income was set off by the rising volume of lending and increasing interest margins on deposits. Costs rose by 3 per cent and the cost ratio improved from 52.2 per cent to 50.0 per cent.

For the medium-sized banking institutions in category B and the small banking institutions in category C, the pre-tax profit in 2006 improved by 33 per cent and 29 per cent, respectively. Net interest income increased by almost 10 per cent as a result of the sustained high lending growth. Net fee income rose by, respectively, 13 per cent and 15 per cent in categories B and C as a consequence of increased activity within asset management and securities trading. In addition, there was rising commission income from the banking institutions' issuance of guarantees in connection with the intermediation of mortgage-credit loans. Positive value adjustments relate primarily to capital gains on equities and gains from the sale of the remaining shares in Totalkredit. The banking institutions in categories B and C have also reported write-downs on loans as net income. This item must still be assumed to be influenced by adjustments to comply with the new accounting rules that entered into force on 1 January 2005.

The small and medium-sized banking institutions show a tendency for staff increases. This contributed to pushing up costs by more than 14 per cent, which will amplify the need to reduce costs again in periods of lower activity.

From 2005 to 2006 the return on equity before tax increased for categories B and C. As Chart 3 shows, this increase is, however, primarily attributable to the banking institutions' gains from the sale of the remaining shares in Totalkredit. Three banking institutions achieved a return on equity excluding Totalkredit gains of less than 10 per cent, while the returns of six banking institutions exceeded 30 per cent.

The Nordic groups all achieved a return on equity before tax of between 22 per cent and 27 per cent. No banking institutions or groups showed negative earnings or returns on equity in 2006.

The development in the return on equity before tax and excluding gains from the sale of shares in Totalkredit from 2005 to 2006 is attributable to the development in four underlying key ratios, cf. Chart 4.

The blue column illustrates the development in the profit ratio, i.e. operating profit before tax over operating income. This key ratio has improved in all three categories. Excluding net reversals of write-downs on loans, only the profit ratios of the Nordic groups have improved. For the banking institutions in categories B and C, costs have risen relatively more than income.

All three categories have seen lower risk-adjusted income, calculated as income as a ratio of risk-weighted items^[2]. Neither interest income nor fee income has increased at the same rate as the risk-weighted items, and particularly the modest development in interest income has a negative impact on the return on equity.

In this chapter, gearing is calculated as assets as a ratio of equity capital. Higher gearing can have a positive impact on the return on equity, but viewed in isolation may be negative in terms of financial stability. The greatest effect of increased gearing was seen for the banking institutions in category C, which was also the category with the highest growth in lending in 2006.

Sustained high growth in lending
Lending growth remained very high in 2006, cf. Chart 5, especially in the category B and C banking institutions, which averaged 25 and 32 per cent, respectively. In category C, more than every fourth banking institution achieved lending growth in excess of 40 per cent. For comparison, lending growth in the Nordic groups in category A was 14 per cent on a portfolio far more diversified in terms of e.g. geographical areas.

In 2004 and 2005, when lending growth escalated, substantial increases were primarily seen in lending to households. The high growth coincided with the banking institutions' introduction of mortgage loans against real property as collateral. In 2006, the rate of growth in corporate lending exceeded the growth rate for total lending to households.

In their financial statements, several banking institutions mention how the high growth in lending to households comprised, inter alia, investment credits, whereby the customer typically has a sum of money that is used as a margin. The banking institution lends the customer an amount that is x times the margin. This money is invested in securities, which are pledged as collateral for the credit together with the margin held by the banking institution. The banking institution thus only incurs a risk if the customer's loss on the investment exceeds the margin. The customer speculates on the return on the investment exceeding the interest payable on the loan, but in order to achieve a positive expected return on the entire exposure, the investment must be placed in more risky assets. Investment credits do not constitute a direct threat to the banking institutions, provided that the investment is monitored closely and divested in time should the market value of the securities fall. However, derived effects may have an indirect impact on the banking institutions. If customers suffer massive losses on the investments, and the activities are discontinued, the banking institutions will see a rapid reduction of exposures, resulting in a diminishing business volume and lower earnings. Moreover, the losses may have a negative impact on the customers' finances and ability to meet payments.

The credit risk is increasing marginally
Danmarks Nationalbank uses its failure-rate model, cf. the chapter on this model (KIM), to calculate a credit-risk measure^[3] that expresses the individual banking institution's expected loss ratio on its lending portfolio. As Chart 6 shows, the credit risk of most Danish banking institutions rose slightly in 2006 compared with 2005, since most dots are above the horizontal line.

The concentration of exposures also affects the credit risk. For example, a lending portfolio can be concentrated on one single sector or geographical area that is dependent on a specific source of income. If the concentration is high, there is an increased risk of substantial losses if precisely this source of income dries up. In this connection it is important that the correlation between the individual concentrations is taken into account, i.e. whether different concentrations of lending are dependent on the same economic conditions, or are mutually dependent, so that losses in one segment will coincide with losses in another segment. The smaller the correlation between risk factors, the lower the risk of substantial losses.

Pursuant to the Danish Financial Business Act, large exposures are defined as exposures that each make up 10-25 per cent of the banking institution's capital base. The sum of these exposures may not exceed an amount equivalent to 800 per cent of the capital base. The calculation of the size of an exposure deviates from the accounting definition in that exposures e.g. include unused credits and loan commitments and are calculated before deduction of write-downs on loans and provisions for guarantees. Groups of interdependent clients are regarded jointly as one exposure.^[4]

To gain an impression of the size of large exposures in relation to the banking institutions' reserves, in Chart 7 the key ratio is stated as a percentage of the excess capital adequacy, i.e. the part of the capital that exceeds 8 per cent.

The ratio of large exposures varies considerably. Among the category 1 banking institutions, the total amount of large exposures as a ratio of the excess capital adequacy has been declining in recent years.

Category 2 stands out with a significantly higher level and far greater spread. Among the category 2 banking institutions, the median for large exposures as a ratio of the excess capital adequacy is almost 450 per cent. Half of the banking institutions in category 2 have reduced their total large exposures as a ratio of the excess capital adequacy, some of them considerably. For two banking institutions, the total large exposures exceed 1,000 per cent of the excess capital adequacy. Among the category 3 banking institutions, the ratio of large exposures to excess capital adequacy has been rising over the last few years, but from a lower level than in categories 1 and 2.

The key ratio does not say anything about the correlation between the individual exposures, but for banking institutions with high concentrations this requires special focus. If most of the large exposures of a banking institution can be influenced by the same economic conditions, e.g. the development in real estate prices, a sudden unexpected dive in real estate prices can have major consequences. Likewise, factors such as geographical location may be important to the banking institutions' assessment of risk.

Increasing deposit deficit
In recent years there has been a structural shift whereby the sector's traditionally large deposit surplus has been replaced by a deposit deficit, cf. Chart 8. The reason is that the growth in deposits has not matched the high growth in lending.

The banking institutions' lending is primarily funded by deposits, which are, on average, an inexpensive and stable source of funding. If deposits are not sufficient to cover lending activities, the banking institutions procure liquidity in the money and capital markets, e.g. via short-term money-market loans and issuance of bonds. This has been the case as the deposit deficit has increased, and in recent years the supply of capital has been ample and inexpensive. However, the large and small banking institutions do not have equal access to the money and capital markets in that the large banking institutions have easier access to the capital markets and can make large issues at lower prices. The small banking institutions are more dependent on access to funding via the money market, but to some extent also issue through larger credit institutions.

Funding through the money and capital markets is generally more sensitive than deposits to changes in the banking institutions' own credit standing. Moreover, changes in the general appetite for risk and risk premiums can rapidly increase the cost of market funding, and the individual banking institution's market funding may potentially " dry out" .

To some extent the deposit deficit reflects the banking institutions' increasing lending against real property as collateral. In the future it will be possible for banking institutions to use these housing loans as collateral for the issuance of covered bonds (SDO). This will provide a new, stable source of funding.

Reduced liquidity reserve
The banking institutions' liquidity reserve can be measured using the Danish Financial Supervisory Authority's liquidity indicators. Among other things, liquid assets must constitute at least 10 per cent of the total debt and guarantee commitments, less the part of the subordinated capital that can be included in the capital base.

Historically, this key ratio has been high for Danish banking institutions, but in recent years the liquidity reserve has been reduced considerably in step with the high lending growth. A number of banking institutions had only very modest liquidity reserves at the end of 2006, cf. Chart 9. Comparison with the situation at end-2001 shows that the number of banking institutions in category B with a liquidity reserve of less than 100 per cent has increased from 5 to 10 of the 12 banking institutions.

Likewise, the number in category C has increased from 12 to 25 of the 37 banking institutions. The average liquidity reserve has almost halved since 2001, to 57 per cent for category B and 87 per cent for category C at end-2006.

The lowest liquidity reserve in one category B banking institution at end-2006 was a mere 0.6 per cent, which is only just above the statutory minimum. The banking institution in question has subsequently obtained liquidity by raising loans and issuing bonds. Of the category C banking institutions, two have liquidity reserves of less than 20 per cent. The low liquidity reserves are remarkable and affect the banking institutions' ability to react to unexpected developments. Tight liquidity may reduce their freedom of action.

The interest-rate risk has declined over a number of years
Interest-rate risk is the risk of losses as a consequence of changes in interest rates in the financial markets and constitutes a major part of the banking institutions' market risk. Interest-rate risk is measured by the Danish Financial Supervisory Authority's key ratio for the share of the core capital that is lost on an increase in the interest rate by one percentage point. In 2006, the average interest-rate risk of the banking institutions in categories B and C was by and large unchanged at, respectively, 2.0 and 2.9 per cent, cf. Chart 10. For both categories, the interest-rate risk has declined substantially over a number of years.

The level of interest-rate risk at the end of 2006 means that on a 1-percentage-point increase in interest rates the banking institutions in categories B and C would overall lose kr. 1.5 billion. This is equivalent to 4.2 per cent of the income for the year in category B, and 7.5 per cent in category C.

Marginally improved capital structure
The high lending growth makes demands of the capital structure of the banking institutions. In categories B and C, the risk-weighted items increased by just over 20 per cent in 2006. Nevertheless, the solvency ratio strengthened slightly in category B and was unchanged in category C, cf. Chart 11, since the capital base was increased in 2006, among other things via the issue of supplementary capital, but also hybrid core capital and share capital.

In the Nordic groups, risk-weighted items increased by 15 per cent in 2006, and these groups have also increased their capital base. The Danske Bank Group issued both share capital and subordinate capital totalling kr. 20.2 billion in 2006 in connection with the financing of its acquisition of Sampo Bank, Finland. This issuance alone affected the solvency ratio of category A by 0.4 percentage point, and the core capital ratio by 0.3 percentage point.

The solvency ratio is the banking institutions' capital including the statutory minimum of 8 per cent. Amounts in excess of the statutory 8-per-cent requirement constitute the banking institutions' reserves in the event of losses that are too large to be covered by earnings. With the introduction of the Basel II Capital Accord, banking institutions must calculate their capital need, which may exceed the statutory 8 per cent, depending on the risks assumed. However, the banking institutions' capital need does not have to be published, and therefore calculations of their reserves cannot take the solvency requirements into account, but must be based on the statutory 8-per-cent requirement for all banking institutions.

Stress tests show unchanged strong resilience
The banking institutions' resilience can be tested in a number of stress scenarios, cf. Table 2. The analyses are static, 1-year analyses based on the result for the year and the capital structure at year-end, and the results thus do not apply to a prolonged period.

The consequences of a decline in net interest and fee income are analysed. The high earnings of the banking institutions in 2006 were achieved against the background of substantial lending growth and high activity in the securities markets. If the general level of activity in the securities markets is falling, the banking institutions' lending for e.g. investment credits must be assumed to be decreasing relatively quickly, and fee income from securities trading will also be affected. In scenario 1, net interest and fee income has therefore been reduced by 10 per cent, so that the income in 2006 would have been at the same level as 1-2 years earlier.

The volume of large exposures has risen considerably in parts of the sector over the last few years. In periods of generally increasing losses on loans, a high concentration risk may further aggravate the situation. In scenario 6, losses have been increased by 10 per cent of the total large exposures. Additional stress elements include rising losses on loans and guarantees, higher interest rates, and the failure of the largest counterparty in the Danish uncollateralised day-to-day money market, which is used by the banking institutions for the exchange of krone-denominated liquidity and management of short-term interest-rate positions.

None of the banking institutions would have suffered a loss in 2006, even if net interest and fee income had been 10 per cent lower, cf. scenario 1. In combination with an increase in the loss ratio by 1 percentage point and a small increase in interest rates, also by 1 percentage point, two banking institutions in category B and seven in category C would have incurred losses, cf. scenario 9.

In the scenarios with rising losses on loans and guarantees, slightly fewer banking institutions in category C would have incurred losses in 2006 compared with 2005. However, three banking institutions in category C would have had negative results even if losses on loans and guarantees had risen by only 1 percentage point, cf. scenario 2. The exposure of the banking institutions in category B to the failure of the largest counterparty bank in the Danish day-to-day money market remained unchanged. On the other hand, the exposures of several category C banking institutions declined. On the loss of 10 per cent of the total large exposures, a total of 10 banking institutions in categories B and C would have incurred losses. One banking institution in category B would even have seen its capital adequacy fall below the statutory 8-per-cent requirement.

Taking the banking institutions' excess capital adequacy into account, Chart 12 shows the number of banking institutions that would have had a solvency ratio below the statutory 8 per cent as the loss ratio on loans and guarantees increased. In 2006, the loss ratio would have had to increase by 3 percentage points for a banking institution to experience problems in meeting the 8-per-cent solvency requirement. Compared with 2005 (the red curve) the resilience is virtually unchanged. Viewed in isolation, the banking institutions' higher lending volumes have had a negative impact on resilience, but this has been set off by rising earnings and the injection of new capital.

Chart 13 illustrates the same development, but in this case the sum of the banking institutions' total assets is shown. On an increase in the loss ratio by 4.5 percentage points, banking institutions with total assets equivalent to 12 per cent of the aggregate total for the Danish banking institutions would have had a solvency ratio of less than 8. The increase in the aggregate assets for categories B and C in 2006 compared with 2005 indicates that these banking institutions' overall market share has increased.

Over the last decades, the highest average loss and provision ratio has been 2.5 per cent, which was in 1994.

Among the Nordic groups, the loss ratio would have had to increase by 1.75 percentage points for one group to lose its excess capital adequacy. In 2005, the equivalent ratio would have to increase by 1.50 percentage points. All six groups would have faced solvency problems on an increase of 4.25 percentage points in 2006, compared with 3.75 percentage points in 2005. The Nordic groups are more immediately exposed to rising losses on loans and guarantees than the other Danish banking institutions. It is important to be aware that the stress test solely indicates the exposure, not the credit risk, of the individual institutions. A substantial proportion of the Nordic groups' lending portfolios comprise mortgage-credit loans, or similar loans, where the probability of losses is very low.

Macro stress test of the Danish banking sector
Chart 14 presents a simple model of losses in Danish banking institutions that can be used to project the sector's aggregate losses on loans and guarantees.

The blue line in the Chart indicates the actual loss development since 1980, while the yellow line shows model calculations of the historical development and the future development. The dashed line indicates the development in losses during a dramatic stress scenario in which GDP falls by 3 per cent and unemployment rises by 5 percentage points over a 3-year period. In this scenario, the banking institutions' losses on loans and guarantees would rise to 1.4 per cent in year 3. This is below the corresponding calculations in Financial stability 2006 because the drop in unemployment gives an improved baseline.

MORTGAGE-CREDIT INSTITUTES

The mortgage-credit institutes are still sound
The income of the mortgage-credit institutes fell marginally in 2006, by 0.3 per cent, reflecting a lower level of activity, while their costs rose by 5 per cent. A low level of new write-downs and the reversal of previous write-downs enabled the mortgage-credit institutes to recognise income of kr. 0.7 billion in 2006. Profit before tax totalled kr. 12.2 billion in 2006, up from kr. 11.7 billion in 2005.

Activity in the mortgage-credit market normalised in 2006 after the record-high level in 2005, which was characterised by a considerable level of remortgaging and redemptions. Gross lending fell from kr. 748 billion in 2005 to kr. 469 billion in 2006, while net lending was at the same high level as in 2005, due to fewer redemptions. Overall, mortgage-credit lending rose by 10 per cent in 2006, compared with 12 per cent the year before. The growth has been particularly strong for capped adjustable-rate loans and fixed-rate loans. As the yield spread between adjustable-rate and fixed-rate mortgage-credit loans has narrowed, it has become less expensive to hedge against rising interest rates. In 2006 the mortgage-credit institutes continued to lose market shares to the banking institutions.

The ability of the mortgage-credit institutes to absorb losses remains extremely good, cf. Chart 15. Their resilience has declined slightly, since the impact from increased lending is not fully set off by an increase in capital base. In 2006 the overall sector would have been able to absorb losses of up to 3.2 per cent of the lending portfolio before the profit for the year and the excess capital adequacy would be undermined. The equivalent figure for 2005 was 3.3 per cent. Actual losses and write-downs for the period were again 0 per cent of lending.

PENSION COMPANIES

The pension companies' returns after tax fell in 2006
After a number of years with increasing returns, the pension companies' returns on investments fell in 2006, cf. Chart 16. Nonetheless, the pension companies generally achieved positive returns. Sound returns on equities and properties were the primary driving forces.

Even though rising interest rates initially lead to losses on the pension companies' investment assets, the long-term effect of a higher level of interest rates is positive for the sector overall. The reason is that on an increase in interest rates the value of the pension companies' guaranteed benefits will probably depreciate more than the investment assets. In other words, the interest-rate exposure (duration) is greater for equity and liabilities than for assets.

Exposure data from 2006 shows that the net balance-sheet effect is positive on an increase in interest rates by 0.7 percentage point, while it is negative on an equivalent fall in interest rates, cf. Table 3. However, a few companies have hedged their interest-rate risk on the asset side to such an extent that the net effect on an increase in interest rates is negative.

The solvency of the pension companies has improved in recent years, partly due to sound returns on investments. All other things being equal, higher solvency makes it possible for the pension companies to take higher investment risks without clashing with the stress scenarios of the Danish Financial Supervisory Authority.

In 2006 the pension companies further increased their portfolios of equities, primarily by reducing portfolios of highly-rated bonds, cf. Chart 17. The proportion of low-rated bonds also rose from 2005 to 2006.

Most pension companies include the results of their stress-test reporting to the Danish Financial Supervisory Authority in their financial statements. At end-2006 none of these pension companies had problems with the red or the more extreme part of the yellow stress-test scenario.