Credit risk

In order to account for non-linear exposures related to a number of structured bonds, we decided to use a Monte Carlo simulation methodology instead of the simpler analytical methods for computing Value at Risk. We employ the same stochastic paths as was used for estimating interest rate risk. Once a month, the simulator generates a set of 1 000 stochastic paths which are then used for different risk management applications (market and credit risk), giving us a coherent picture of our risk profile.

For the monthly reports we set the time horizon to one year because we are more interested in longer-term effects. This fairly long VAR horizon is another reason why we prefer a simulation approach over an analytical model.

However, this time feature also makes the assumption that the portfolio is not changing between now and the VAR horizon, quite questionable.³

A future enhancement of the model could be to have quick marginal and incremental analyses available, similar to correspondent functions of the long-term credit risk analysis tool (see below).

of swaps can fluctuate significantly. Second, the credit standing of a financial partner can deteriorate significantly over the years.

Short term risk

During the year, a quite stable amount of cash is invested in the national and international money markets. These investments are restricted by the limits set by the supervisory board, with either individual limits for specific banks or general limits for potential partners in the commercial paper market.

When Austrian Treasury Bills in a foreign currency are issued, we often hedge the short term forex risk by using a forward. As a result, the potential value of such a forward reduces the available limit of the counterparty in question. The reporting is done on a monthly basis and is controlled automatically by the treasury software. This means that the treasury trader has to check simultaneously whether or not the limit has been reached before negotiating with a counterparty bank.

Long term risk

The swap portfolio is the main source of long term credit risk created by debt management. In the second half of the 1990s, AFFA began to pay more attention to this type of risk, as it became clear that credit risk arising from swaps can lead to substantial losses. Fortunately we were never forced to realise losses caused by defaulted derivative partners. Nevertheless, in 2000 AFFA’s supervisory board imposed new guidelines for managing credit risk.

The guidelines distinguish the following three types of exposure:

● Current Exposure.

This is the exposure on the valuation date. It is either the current non-negative value of the transaction or zero (i.e.when the value of the transaction is not positive).

● Potential Exposure:

Current exposure is determined by market factors. These market factors change and therefore also the exposure. Potential exposure shows the maximum value exposure can reach on a certain date in the future with a well-defined probability.

● Peak Exposure:

To obtain a full credit risk profile of the counterparty, the highest Potential Exposure over the whole lifetime of all long term transactions with the counterparty is calculated. This is called “Peak Exposure”. This risk measure forms the basis for all long term credit limits.

Peak exposures are calculated by generating the standard set of stochastic paths using Monte Carlo procedure. The exact methodology for determining peak exposures is dependent on the existence of a legally enforceable close-out netting agreement with a particular counter-party, or, even better, on the existence of a collateral agreement between AFFA and that counterparty. Both close-out netting and collateral agreements lower credit risk and are therefore taken into account in our credit risk management system by lowering peak exposures, but leaving the credit limits of the counterparties unchanged.

The guidelines establish two layers with credit limits. First, there is a limit for each individual counterparty, and second, there is also a limit for certain groups of counterparties. For individual counterparties, peak exposure and time to maturity have limits. For groups of counterparties, the guidelines limit the share of each group of counterparty in the sum of all counterparties’ peak exposure.⁴

When setting individual limits, we start with a maximum limit and then make deductions for smaller counterparties and for those with lower creditworthiness. The size of each deduction should reflect the credit risk which we assume can be estimated by taking counterparties’ ratings from Moody’s and Standard & Poor’s and by using shareholders’ equity as a measure of size. This means that large counterparties with high credit ratings will get the full limit, while the others receive lower limits. Front office dealers are required to check these limits before they enter into a swap. Determining the counterparty’s new peak exposure (that is, including the new swap) is a

non-Figure 7.4. Current exposure in previous year

1 400 1 200 1 000 800 600 400 200 0

AAA Millions of EUR

AA A < A Collateral

Mar. 04 Apr. 04 May 04 June 04 July 04 Aug. 04 Sep. 04 Oct. 04 Nov. 04 Dec. 04 Jan. 05 Feb. 05

trivial calculation because it depends not only on the characteristics of the new swap, but also on those of all the other swaps with that counterparty in the portfolio. A new swap can either increase the peak exposure, when it is positively correlated with the other swaps, or decrease the peak exposure when it is negatively correlated. A proper risk analysis would require to run the 1 000 simulation paths of the long term exposure analysis involving the whole portfolio of swaps with that counter-party in question. This could take hours if it is an important counterparty with many swaps in our books.

Therefore, for intra-day business, we use a quicker approximation procedure and recalculate the whole portfolio overnight in the proper way. If the calculation shows that the actual use of the credit limit for a particular counterparty is low, then the approximation will not create a serious problem.

But more caution must be exercised when actual use is close to the limit.