The Financial Sector

Again in 2005 the Danish banking institutions and the Nordic financial groups posted record-high earnings. Growing business volumes with strong lending growth, intense activity in the securities markets, as well as remortgaging activity generated higher fee income. This was also yet another year with historically low write-downs on loans due to continued favourable economic conditions and new accounting rules. Finally, the banking institutions' securities portfolios yielded substantial capital gains.

Overall the banking institutions and the Nordic groups still appear to be robust. The strong lending growth and a minor erosion of capital adequacy mean that the resilience of medium-sized and small banking institutions to increasing losses has decreased a little. At the same time, there is a certain tendency for the credit risk on the lending portfolio to be higher for the banking institutions with strong growth in lending. The resilience of the Nordic groups to losses has also declined a little as a result of high lending growth.

The strong growth in lending makes it important that the banking institutions, in their risk and capital management, allow for both direct and indirect effects of a possible cyclical reversal, as well as the fact that under the new accounting rules write-downs on loans are not based on the prudential accounting principle, and thus cannot serve as a buffer.

THE SIGNIFICANCE OF FINANCIAL INSTITUTIONS TO FINANCIAL STABILITY

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

The mortgage-credit institutes are providers of credit for financing of real estate, which makes them the largest bond issuers in Denmark. The development in the mortgage-credit institutes can affect the banking institutions' earnings directly via ownership and indirectly via competition in the home-financing market.

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

NORDIC GROUPS AND DANISH BANKING INSTITUTIONS

Earnings and return on equity
Both the Nordic groups in category A and the Danish banking institutions in categories B and C posted record-high results in 2005. The categories are described in Box 1. The pre-tax profit of the Nordic groups totalled kr. 86.3 billion, i.e. an increase of 23 per cent on 2004, and the return on equity after tax rose to 18.5 per cent^[1] , cf. Table 1. Categories B and C recorded similar earnings increases, by 35 per cent and 29 per cent, respectively, and the return on equity after tax is 17.4 and 14.5 per cent, respectively.

Net interest and fee income was positively influenced by increasing business volumes, with strong lending growth, increased securities trading and continued high remortgaging activity. Fee income still accounts for an increasing proportion of total net interest and fee income, i.e. just over 30 per cent. Capital gains more than doubled for the Nordic groups and increased by 31 per cent for category B and 12 per cent for category C.

The costs of the Nordic groups rose by 4 per cent as a result of expansion into new business areas, among other factors. The cost ratio, however, was reduced from 55.4 per cent in 2004 to 52.2 per cent in 2005. The banking institutions in categories B and C recorded cost increases of 9 and 8 per cent, respectively, but the cost ratio was reduced to just under 55 per cent.

In 2005 write-downs on loans and guarantees for the year were reduced from an already very low level, and the Nordic groups could report a total amount of kr. 1.2 billion as income in 2005. The new accounting standards, cf. Box 2, are applied to the calculation of write-downs in 2005, which impedes historical comparison.

The return on equity before tax is increasing for all three categories, cf. Chart 3. For a few institutions the return on equity before tax exceeds 30 per cent, and for almost half of the banking institutions in categories B and C the return on equity exceeds 20 per cent. Only two of the banking institutions represented show a return on equity of around 10 per cent. For all the Nordic groups the return on equity before tax exceeds 20 per cent. The return on equity can be broken down into a number of key ratios, cf. Box 3.

For the Nordic groups in category A, the high return on equity is achieved through cost efficiency and high equity gearing, cf. Table 2. The high gearing reflects the groups' focus on their capital structures in order to generate returns for the shareholders and reduce overcapitalisation.

The groups have a relatively low risk level compared to the banking institutions in categories B and C, which is attributable to the high proportion of lending against real property as collateral. In addition, the banking institutions in categories B and C offer mortgage-credit loans in collaboration with mortgage-credit institutes, typically providing a guarantee against losses to the mortgage-credit institute. These guarantees are not included in the total assets, but in the risk-weighted items, and the effect of the guarantees is particularly pronounced in category C which has a risk level of 109 per cent in 2005. On the other hand, the gearing is low for institutions in category C compared to category B and the Nordic groups.

The banking institutions in categories B and C are by and large at the same level, but lower than the Nordic groups, in terms of the profit ratio, i.e. operating profit over operating income. Recent years have seen a tendency for many small and medium-sized banking institutions to focus on growth rather than reducing costs.

Chart 4 shows how the increase in return on equity has been achieved over the last four years. For example, the higher gearing contributed 0.9 percentage point of the total increase of 2.5 percentage points in the return on equity for the Nordic groups from 2004 to 2005, and the improved profit ratio contributed 2.4 percentage points, whereas a decrease in the risk level reduced the return on equity by 0.9 percentage point.

The banking institutions in categories B and C have increased their risk level in recent years, particularly from 2004 to 2005, and at the same time risk-adjusted income has declined. As a result, the banking institutions' income per "krone of risk" is lower, or, in other words, they charge less for the higher risk.

All three categories have seen strong growth in the profit ratio. The low level of write-downs on loans contributed significantly to the growth in income, cf. Chart 5. This effect is particularly pronounced for the banking institutions in categories B and C, but also among the Nordic groups. The low write-downs are attributable to favourable economic conditions and reversals of previous years' provisions. The latter must be considered a transitional effect in connection with the implementation of new accounting standards. Other income, e.g. remortgaging fees and value adjustments, may also be of a one-off nature. At the same time, an increasing cost level due to e.g. establishment of branches or staff expansions may contribute to dampening profits in future, if the income base is reduced. In addition, expansion of the branch network may intensify competition.

Lending growth
Growth in lending continued to increase in 2005 to 29 per cent for both categories B and C, cf. Chart 6. This is the highest level since the mid-1980s. 10 per cent of the banking institutions are experiencing lending growth in excess of 45 per cent. In comparison, lending by the Nordic groups rose by 14 per cent in 2005.

The strong growth in lending by the banking institutions in recent years is driven by the sustained low level of interest rates and rising property prices. Growth in lending to households accounts for almost half of total lending growth, and 60 per cent of the household element stems from lending for home-financing purposes. The banking institutions have gained market shares from the mortgage-credit institutes by offering mortgage loans ("prioritetslån", loans collateralised by real estate). Often the excess liquid proceeds for the customer are placed in a deposit account subject to the same interest terms as the loan. Against this background, a proportion of the banking institutions' lending growth is reflected in increased deposits.

In a historical perspective, periods with high lending growth have been followed by gradual deterioration in the credit quality, with increased losses and provisions in subsequent periods. cf. Chart 7. It is by no means certain that history will repeat itself, in view of the increased focus on management of credit risk, e.g. the implementation of the new capital-adequacy rules, Basel II. Under Basel II the capital requirement can, however, be more sensitive to fluctuations in credit quality, and the banking institutions should be aware of this and adjust their capital bases accordingly.

Capital structure
The solvency ratio for banking institutions in categories B and C is still considerably above the statutory 8-per-cent requirement, but the strong growth in lending in the two categories has increased the risk-weighted assets, with an adverse impact on the core capital and solvency ratios. As a countermeasure, the banking institutions in category B have primarily issued hybrid core capital^[2] (loan capital that may be included in the core capital according to certain rules), and the banking institutions in category C have primarily raised subordinated debt. As a result, the core capital and solvency ratios in category B are almost unchanged, while especially the core capital ratio for category C has been reduced, albeit from a high level, cf. Chart 8.

The average solvency ratio for the Nordic groups in category A is just over 10 per cent. The groups have primarily raised subordinated debt, but also hybrid core capital, so that the solvency ratio has remained almost unchanged. The capital structure has been actively adjusted by repurchase of own shares, and the core capital ratio has been reduced to just over 7 per cent on average.

An increase in the banking institutions' losses will first and foremost reduce earnings, and any additional losses are to be borne by the buffers, i.e. the proportion of the capital in excess of the statutory 8-per-cent requirement. Excess capital adequacy has been reduced from 4.9 per cent in 2004 to 4.6 per cent at end-2005, cf. Chart 9. Under the previous accounting rules based on the prudential principle the banking institutions' loan loss provisions were not necessarily reflected in subsequent losses whereby the accumulated provisions partly served as a buffer. Under the new IFRS accounting standards as from 1 January 2005, the banking institutions may not write down loans and guarantees unless there is an objective indication of impairment - the principle of neutrality. This means that any write-down will most likely be associated with a loss. Thus, under the new rules, accumulated write-downs will not have the same buffer value as was the case for accumulated provisions. This implies erosion of the banking institutions' total buffers, cf. Chart 9.

Market assessment of the Nordic groups
Prices in the financial markets contain information on investor expectations of the future. On the basis of equity prices and accounting data, it is possible to estimate a market-based risk measure – distance to insolvency^[3] . This measures the changes (in number of standard deviations) in market asset value that can be accommodated within the group's capital buffers, write-downs and earnings. Chart 10 shows the distance to insolvency for the Nordic groups.

For each Nordic group the distance to insolvency is more than three standard deviations. Consequently, the market assessment is that the group's buffers are at least three times greater than the expected changes in market asset value. Distance to insolvency may be seen as a value-at-risk measure, with a distance to insolvency of 3 standard deviations corresponding to a market assessment of a mere 0.13 per cent probability^[4] that a banking institution's losses will exceed its buffers. Overall, the distance to insolvency is greater than previously and is at a level indicating a market assessment that the risk of a group's non-compliance with the statutory solvency requirement is very low.

RISK FACTORS FACING DANISH BANKING INSTITUTIONS

Assessment of banking institutions' credit risks
Credit risk is the risk of a banking institution suffering a loss should customers or other counterparties default on their obligations to that banking institution.

Danmarks Nationalbank's failure-rate model can be used to analyse the credit risk on the lending portfolios of Danish banking institutions, cf. Box 5. For each banking institution a credit-risk measure can be estimated. The measure reflects the proportion of the lending portfolio that the banking institution may expect to lose over 1-2 years. The credit-risk measure is used to rank the banking institutions according to the degree of credit risk on their lending portfolios.

The credit risk is generally lower for the large Danish banking institutions than for other Danish banking institutions, cf. Chart 11. The principal reason is that the estimated failure rate for the companies using the large Danish banking institutions is, on average, less than half the estimated failure rate for the companies using the small and medium-sized banking institutions. This can be attributed to such factors as a higher return on assets, the fact that the companies are larger and older, as well as higher solvency and a lower debt ratio^[5] . In contrast, banking institutions in categories 2 and 3 have a higher proportion of lending to private customers, at around 50 per cent compared to 25 per cent for the large banking institutions. Viewed in isolation, this implies a lower credit risk since the probability of losses on households is generally lower than the probability of losses on companies. Overall, for the banking institutions in categories 2 and 3 the credit risk is higher and the dispersion in the credit-risk measure is greater.

A comparison of the banking institutions' credit risk with the growth in lending shows that high lending growth has a certain tendency to coincide with a higher credit risk, cf. Chart 12. At the same time, the dispersion in the credit-risk measure is relatively high among banking institutions with high lending growth.

Money-market exposure
The Danish money market is the market for krone-denominated exposures among banking institutions and mortgage-credit institutes. This market is used primarily for exchange of krone-denominated liquidity and management of short-term interest-rate positions.

The volume of exposures among Danmarks Nationalbank's monetary-policy counterparties in the uncollateralised day-to-day money market rose by 35 per cent to kr. 10.4 billion on average per day in 2005. The market was thus characterised by ample liquidity and many positions at the short end of the yield curve. The market scale fluctuated considerably from day to day, with total lending in a single day peaking at kr. 26.8 billion and bottoming out at kr. 3.2 billion.

Table 3 shows the average exposure among credit institutions in the uncollateralised day-to-day money market in 2005, calculated on the basis of transactions in Danmarks Nationalbank's payment system, Kronos.

In 2005, the average daily exposure of the Nordic groups in category A amounted to kr. 943 million, corresponding to an increase of 32 per cent on 2004. The groups are by far the largest players in the market, with a market share of 77 per cent in 2005. The market share of foreign credit institutions, excluding Nordic groups, decreased from 12 per cent in 2004 to 8 per cent in 2005. Average lending conceals considerable dispersion over days and between credit institutions. Table 4 shows the highest average exposure per credit institution in the categories in 2004 and 2005. The highest exposure for the Nordic groups in category A increased from kr. 3.0 billion in 2004 to kr. 3.8 billion in 2005.

Interest-rate risk
The interest-rate risk of the banking institutions is measured in terms of the proportion of the core capital (tier 1 capital) that is lost on an upward parallel shift in the yield curve of 1 percentage point.

The banking institutions in category B have almost halved their interest-rate exposure, from 4.1 per cent in 2001 to 2.2 per cent in 2005, cf. Chart 13. The reduction has taken place in step with the decline in the general level of interest rates and may reflect expectations of rising interest rates. The dispersion among the banking institutions is great, however. Among the institutions in categories B and C, 10 per cent recorded an interest-rate risk of more than 4.6 per cent and 6.2 per cent, respectively at end-2005.

Stress testing
The resilience of the banking institutions to a number of hypothetical scenarios can be tested. The consequences of an interest-rate increase, higher losses on loans and guarantees and the failure of the largest counterparty in the uncollateralised day-to-day money market during the year are analysed. The analyses are static and cover a period of one year. They are based on the profit for the year and capital structure at year-end. Thus, no allowance is made for the possibility that the banking institutions could have responded to the changed conditions, so the results of the stress tests are not suitable for analysing the resilience in the longer term. The scenarios selected in Financial stability 2005 still apply.

The International Monetary Fund, IMF, is performing a macro stress test of the Danish banking sector in 2006, cf. Box 6.

Slightly fewer banking institutions in categories B and C would show a loss in the event of higher losses on loans and guarantees in 2005 than in 2004, cf. Table 5. Viewed in isolation, increasing lending reduces the resilience to higher losses, but this effect is offset by higher earnings. The results for the scenarios with interest-rate increases also reflect the higher earnings concurrently with a reduced interest-rate risk.

In scenario 4, the banking institutions are assumed to suffer a loss on the failure of the largest counterparty in the uncollateralised day-to-day money market. This scenario also shows that the banking institutions were more resilient in 2005, despite the general increase in exposures.

In scenario 8, the loss ratio is increased by 2.5 percentage points and the interest rate is increased by 3 percentage points. In this scenario, all banking institutions in categories B and C would have recorded negative results.

Taking the banking institutions' excess capital adequacy into account, Chart 14 shows how many banking institutions would have a solvency ratio below 8 as the loss ratio on loans and guarantees increases.

Earnings and excess capital adequacy are not lost until losses rise by 4 percentage points so that the solvency ratio of one banking institution in category B falls below 8 per cent. The strong lending growth and declining solvency imply slightly lower resilience for the banking institutions in categories B and C taken as one, since more banking institutions encounter solvency problems sooner when losses on loans and guarantees are rising (the red curve extends further to the southeast than the bars).

Chart 15 shows the same situation, but the banking institutions are included with their total assets. This gives an idea of the size of the affected banking institutions.

As far as the Nordic groups are concerned, losses must rise by 2.75 percentage points before two groups fall below the solvency requirement. The corresponding figure in 2004 was 3 percentage points. All six groups would encounter solvency problems on an increase in losses by 4.25 percentage points, compared to 4.75 percentage points in 2004.

Charts 14 and 15 show only the size of the losses that the banking institutions can sustain, but not the probability of the losses occurring. The probability of losses for the individual banking institution depends on the credit risk on the lending portfolio, among other factors. If the credit risk on a banking institution's lending portfolio is relatively high, stronger resilience to rising losses should be sought. Chart 16 compares the resilience from Chart 14 with the probability of losses expressed by the credit-risk measure. There is no unequivocal positive connection between resilience and risk of losses, so it is not necessarily the banking institutions with the highest credit risk that can sustain the highest losses.

MORTGAGE-CREDIT INSTITUTES

In 2005 the losses and write-downs on loans of the mortgage-credit institutes were influenced by considerable reversal of previous provisions. Together with the very low write-downs in 2005, this enabled the mortgage-credit institutes to report net income of kr. 0.5 billion.

The mortgage-credit market was characterised by competition on products. Gross new lending was record-high at kr. 750 billion in 2005 as a result of the extensive remortgaging to capped loans and deferred-amortisation loan, as well as extensive remortgaging activity. Concurrently, new construction and price increases for owner-occupied homes contributed to 12 per cent growth in mortgage-credit lending. In contrast, the banks' mortgage loans have gained business from the mortgage-credit institutes. The remortgaging activity and the shift towards new products prompted a large number of early redemptions, i.e. just over kr. 540 billion in total. Net lending totalled just over kr. 160 billion.

The income of the mortgage-credit institutes rose by 19 per cent to kr. 15 billion in 2005, with increasing contributions accounting for approximately half. Costs rose by 4 per cent due to the higher activity. Profits before tax increased by 29 per cent in total to kr. 12 billion.

The mortgage-credit institutes' resilience to losses is still high and growing, cf. Chart 17. In 2005 the sector as a whole could bear losses of up to 3.2 per cent of the lending portfolio before the profits for the year and the excess capital adequacy would be lost. Actual losses and write-downs for the year accounted for 0 per cent of the lending portfolio.

PENSION COMPANIES

The Danish pension companies posted record-high returns on investments in 2005, cf. Chart 18. Rising equity prices combined with falling interest rates enabled the companies to achieve a positive return on both equity and bond portfolios. More than half of the pension companies recorded double-digit returns after tax, which has not been observed since 1999.

In view of the favourable development in returns, today the sector is robust. At end-2005, only few pension companies experienced problems with the yellow-light stress test of the Danish Financial Supervisory Authority, while no pension companies recorded problems with the red-light stress test. The sector's increased use of financial derivatives, which provides for a better match between the interest-rate risk on assets and liabilities, respectively, contributed to this development.

Looking ahead, the low level of interest rates still constitutes a significant replacement risk for the pension companies since old 4.5 per cent guarantees continue to weigh heavily on the liabilities of a number of pension companies^[6] In view of the current level of interest rates, it can be difficult to pay an annual rate of return after tax of 4.5 per cent, cf. Chart 19.

In 2005, the pension companies reduced the bond element of total investment assets, primarily to increase the equity share, cf. Chart 20. The share of other financial assets is also higher, reflecting e.g. extended use of interest-rate derivatives and other financial derivatives.

[1] In comparison, the return on equity after tax for banks in the euro area was 10.5 per cent in 2004. For the largest banks in the euro area the return on equity was 13.6 per cent in 2004, and preliminary figures for the first six months of 2005 show a return after tax of 20.8 per cent, cf. ECB, Financial Stability Review, December 2005.

[2] Hybrid core capital is described in more detail in: Financial stability 2005 and Bundgaard, Birgitte and Suzanne Hyldahl, Structure of the Banks' Capital – New Statutory Requirements and Opportunities, Danmarks Nationalbank, Monetary Review, 3rd Quarter 2002.

[3] See the description of methodology in the chapter on market-based risk measures in Financial stability2004 and the chapter on analysis of bank equity prices in Financial stability2005.

[4] 0.13 per cent corresponds to the probability mass in a normal distribution for incidents of more than 3 standard deviations.

[5] See also the description of the failure-rate model in Financial stability 2004.

[6] As from 1982 until mid-1994 the pension companies were free to guarantee the pension savers a minimum annual return after tax of 4.5 per cent over the life of the pension. In 1994, the Danish Financial Supervisory Authority lowered the rate to 2.5 per cent, followed by a further lowering, to 1.5 per cent, in 1999.