Risk Management of Central-Government Debt

Interest-rate risk related to central-government debt is managed via a strategic benchmark for the duration of the debt portfolio. The target band for the duration in 2006 remains unchanged at 3.0 years ± 0.5 years.

The foreign government debt is primarily denominated in euro. As a result of Denmark 's fixed-exchange-rate policy vis-à-vis the euro, the exchange-rate risk is limited.

Credit risk related to the central government's swap portfolio is limited via requirements of the credit ratings of counterparties, as well as unilateral collateral requirements. At end-2005, 98 per cent of the central government's swap portfolio, measured in terms of principal, was covered by unilateral collateral agreements.

Operational risk is limited via the use of standardised financial instruments, clear and unambiguous procedures and a clear division of functions in Government Debt Management .

8.1 INTEREST-RATE RISK

Interest-rate risk is the risk of higher interest costs on the central-government debt as a result of the development in interest rates. Interest-rate risk on the debt is managed on a consolidated basis, cf. Box 8.1.

The key strategic benchmark for the central government's interest-rate risk is the duration of the debt portfolio, cf. Box 8.2 . Duration is an expression of the average fixed-interest period on the debt portfolio and thus a summary measure of the trade-off between costs and risk. The longer the duration, the lower the interest-rate risk on the central-government debt. On the other hand, long duration will typically be associated with higher expected interest costs as the level of interest rates normally increases with the term to maturity.

Interest-rate risk, and thus duration, is affected by the issuance and buy-back strategy, as well as by interest-rate swaps. The issuance and buy-back strategy is primarily aimed at building up liquid government bonds by issuing new securities and buying back older series.

The issuance strategy is separated from the management of the interest-rate risk by means of interest-rate swaps and by targeting buy-back of certain bonds.

Interest-rate swaps are used to transfer interest-rate exposure from one maturity segment to another. Buy-backs typically take place in securities maturing within the next few years, so that the interest-rate fixing is reduced in these years and increased in later years. The interest-rate fixing is the amount for which a new, unknown rate of interest is to be fixed within one year, cf. Box 8.2 . The procedure contributes to reducing the central government's exposure to fluctuations in the level of interest rates.

Every year, the duration level of the government debt in the subsequent year is determined. The decision is taken on the basis of a long-term analysis of the development in the interest-rate fixing of the central-government debt, as well as simulations using the Cost-at-Risk (CaR) model. The CaR model is used to quantify expected interest costs and risks subject to various assumptions of the duration and the interest-rate fixing. The analysis of the interest-rate fixing indicates the development in the duration, while the CaR analysis illustrates the cost and risk profile for various strategies.

In 2005, the strategic benchmark for the Macauley duration was 3 years ± 0.5 years, and 3 years ± 0.25 years for the duration measure calculated with a fixed discount rate and a fixed balance of the central-government account, cf. Chart 8.1.1. In view of the low level of interest rates, it was decided to keep the duration in the upper half of the band in 2005, reducing the need for interest-rate swaps.

Analysis of the interest-rate fixing for the debt portfolio, 2006-12
The interest-rate fixing in a given year is an expression of the central government's exposure to fluctuations in interest rates. All other things being equal, a general increase in the level of interest rates by 1 percentage point will increase the interest costs in that year by 1 per cent of the interest-rate fixing.

The duration of the debt portfolio does not include any information on the absolute interest-rate exposure or its dispersion over time. Consequently, an analysis of the development in the interest-rate fixing of the debt portfolio is used to support the choice of duration.

In 2005, the interest-rate fixing was around DKK 260 billion, of which approximately DKK 190 billion comprised Treasury bills and interest-rate swaps. Maintaining this level will reduce the central government's interest-rate exposure relative to GDP over the coming years. The interest-rate fixing relative to GDP expresses the central government's real exposure. In 2005, the interest-rate fixing was 17 per cent of GDP. Alternatively, the real exposure can be maintained by increasing the interest-rate fixing with GDP growth.

The basis for the analysis of the development in the interest-rate fixing is a technical projection by the Ministry of Finance that shows sustained budget surpluses. If the interest-rate fixing is maintained at the current level, or increased with GDP, a larger share of the debt portfolio will be converted to floating interest rates. This results in ongoing reduction of the duration, cf. Chart 8.1.2.

In summary, the analysis of the development in the interest-rate fixing indicates that the duration of the central-government debt can be gradually reduced over the coming years without any increase in the central government's interest-rate exposure. Assuming an alternative course for the debt reduction and GDP, the conclusion of the analysis may change.

Cost-at-Risk analysis of the duration target for 2006
The CaR model is used to illustrate the trade-off between interest costs and interest-rate risk on the central-government debt. In the CaR model, the interest costs on the government debt are simulated 10 years ahead on the basis of 2,500 scenarios for the development in the level of interest rates. Based on these scenarios, expected interest costs and various risk measures are calculated, cf. Box 8.3.

The simulation is based on the existing debt portfolio, technical budget projections by the Ministry of Finance, a strategy for future borrowing, and a strategy for the use of interest-rate swaps. By varying the swap strategy in the model, it is possible to analyse various courses of the duration and interest-rate fixing of the debt portfolio.

The CaR analysis supplements the analysis of the development in the interest-rate fixing, which is solely a measure of exposure. The actual risk depends on the uncertainty concerning interest rates, and e.g. unchanged interest-rate fixing may entail a higher risk if the interest-rate volatility increases.

Against the background of the above results, which point to a reduction of the duration in the coming years, the consequences of reducing the duration in 2006 have been analysed in CaR. The starting point of the analysis is two strategies, where duration is set at 2.5 years and 3.0 years, respectively, in 2006, cf. Chart 8.1.3. As from 2007, the interest-rate fixing is kept at the 2006 levels for the two strategies. This method isolates the impact of the duration decision for 2006 since the two strategies differ solely in terms of the number of interest-rate swaps in 2006.

The Ministry of Finance's technical projection of the government budget balance shows that the central-government debt will decrease. This helps to explain the falling trend for interest costs over the simulation horizon. A simulated increase in the general level of interest rates has the opposite effect. The impact of the reduction in the government debt dominates in the projection, which contributes to explaining why both expected interest costs and absolute CaR decline over the simulation horizon.

If the duration is shortened by 0.5 years in 2006, the expected interest costs will, all other things being equal, be reduced by around DKK 100 million in 2006 and by around DKK 190 million in 2007. The risk, measured as absolute CaR, remains virtually unchanged in 2006 and increases by more than DKK 250 million in 2007. The reason is that interest-rate swaps transacted in 2006 are not fully reflected in the risk profile until 2007.

Over longer horizons, the risk increases by up to around DKK 500 million in 2010, while the expected savings are on the low side of DKK 100 million.

The expected savings on a reduction of the duration in 2006 are modest in relation to both the total interest costs and the increased interest-rate risk. Firstly, the yield curve is rather flat, cf. Chart 8.1.4, and thus the saving on short-term relative to long-term borrowing is limited. Secondly, the level of interest rates is lower than the average interest-rate level over the simulation horizon, which contributes to reducing the central government's expected net saving on interest-rate swaps transacted in 2006. The fixed swap rate received by the central government throughout the lifetime of the swap reflects the level of interest rates at the time of its transaction, while the floating rate is set on market terms on a current basis.

In view of the above, it has been decided to maintain the strategic benchmark for duration unchanged at 3 years ± 0.5 years in 2006. If the development in interest rates leads to a change in the trade-off between short-term and long-term borrowing relative to the CaR assumptions, the duration can be adjusted within the duration band.

8.2 EXCHANGE-RATE RISK

Exchange-rate risk is the risk that the value of the central-government debt in kroner increases as a result of changes in exchange rates. On a consolidated basis, cf. Box 8.1 , the foreign government debt is exposed solely in euro. This entails a low exchange-rate risk due to Denmark 's fixed-exchange-rate policy vis-à-vis the euro. In addition, Danmarks Nationalbank's foreign-exchange reserve is predominantly exposed in euro.

8.3 CREDIT RISK

Credit risk is the risk of financial loss as a consequence of a counterparty's default on its payment obligations. Interest-rate and currency swaps entail credit risk for the central government. The credit risk arises because the market value of a swap may develop to the advantage of the central government during the lifetime of the swap. When a swap is transacted, its market value is normally zero. Over time, the development in interest and exchange rates will entail that the market value may become both positive and negative for the central government. A swap with a positive market value is an asset for the central government and is thus subject to credit risk since the central government is exposed to the swap counterparty's ability to pay. The key principles of the central government's credit management are described in Box 8.4 . A more detailed account is presented in the Appendices.

In 2005, the central government transacted 32 new swaps with a total principal of DKK 17.2 billion, while 10 swaps expired. At end-2005, the central-government swap portfolio comprised 351 swaps, with a total principal of DKK 148.7 billion, cf. Table 8.3.1.

The market value of the swap portfolio
The development in the market value of the central government's swaps primarily reflects fluctuations in the level of interest rates and in the dollar rate. The market value, and thus the credit exposure, increases when the dollar appreciates and/or interest rates fall. The reason is that the central government receives dollars in some currency swaps and mainly receives long interest on the interest-rate swaps.

The exposure of the swap portfolio to fluctuations in the dollar rate has been considerably reduced, cf. Table 8.3.2. The reason is partly that some currency swaps have expired on which the central government has received dollars, and partly that the central government transacts currency swaps from kroner to dollars in connection with re-lending to Danish Ship Finance A/S. Since the central government pays dollars in these swaps, the net exposure to fluctuations in the dollar rate is reduced.

The market value of the central government's swap portfolio rose substantially in the 1st half of 2005 in view of the falling level of interest rates, cf. Chart 8.3.1.

Subsequently, the market value has declined by around DKK 3 billion, reflecting the rising interest rates in the autumn, to its present level of almost DKK 8 billion, cf. Table 8.3.3.

Credit exposure on the swap portfolio
The credit exposure on the swap portfolio is calculated on the basis of the current market value of the portfolio, the value of pledged collateral and a supplement to take account of potential future fluctuations in the market value.

New swaps are transacted only with counterparties that have signed a unilateral collateral agreement. In 2005, the credit exposure on the swap portfolio rose by DKK 0.6 billion to DKK 6.5 billion, cf. Table 8.3.4. This is attributable to an increase in the current exposure by DKK 0.6 billion, which was only partly offset by an increase of DKK 0.2 billion in the pledged collateral, and an increase of DKK 0.2 billion in the potential exposure.

At year-end, the central government had signed unilateral collateral agreements with 22 counterparties. Swaps transacted with these counterparties account for 98 per cent of the total swap portfolio in terms of swap principals, cf. Table 8.3.5. The explanation for the coverage of 0 per cent in the A- rating class is that the central government previously transacted a swap with a single counterparty that is now in this rating class and with which no collateral agreement has been concluded, cf. below. For the highest rating class, the proportion covered is also somewhat lower, but on the other hand the credit risk is limited in view of the high credit ratings of these counterparties. The proportion of the swap portfolio not covered by collateral agreements decreases as old swaps with counterparties that have not signed collateral agreements expire.

The distribution of the swap portfolio by counterparties shifted somewhat in 2005, cf. Chart 8.3.2. During the year, three counterparties were upgraded and three downgraded by either Moody's, Fitch or Standard & Poor's. In five of these six cases (three downgradings and two upgradings), the changed rating led to a change of line (lines are based on the lowest rating given to a counterparty by either Moody's, Fitch or Standard & Poor's).

The central government has very small exposures to the three downgraded counterparties, including its only counterparty in rating class A-, since none of these has concluded a collateral agreement. Consequently, these banks are no longer used as counterparties – and have not been for some years – and the central government's largest credit exposure with one of the downgraded counterparties is DKK 52 million on a swap maturing in 2007.

Two of the downgraded counterparties are German Landesbanks, and their downgrading is a result of the discontinuation of guarantees by the German Länder. Existing debt obligations are still covered by guarantees until they mature, and the rating of these liabilities is therefore unchanged, cf. for example the rating agency Standard & Poor's.^[1] The central government's transactions with the Landesbanks were concluded prior to the discontinuation of the guarantees.

The prudent accounting principle entails that the Landesbanks are registered in the central government's credit-management system on the basis of their new, lower ratings.

8.4 OPERATIONAL RISK

The Bank for International Settlements (BIS) has defined operational risk as "…the risk of loss resulting from inadequate or failed internal processes, people and systems, or from external events".^[2]

Operational risk is minimised by applying a number of different measures. Government Debt Management is divided into front, middle and back offices with separate functions. A clear division of functions reduces operational risk and facilitates internal control. Moreover, only standardised, well-known financial instruments are used, and legal risk is minimised by exclusively using standardised contracts.

Clear procedures have been defined for the individual tasks, and all procedures are peer-reviewed and approved by the manager in charge. In 2005, Government Debt Management's procedures were transferred to a database where review and approval take place electronically in a defined workflow.

In the event of major business disruptions, a contingency plan has been prepared, including a second site from which key Government Debt Management activities can be continued.