CREDIT LOSSES
Credit exposure indicates, with a certain probability, the credit losses. The measurement of exposures and credit losses plays a key role in the pricing of the contract, that is, it defines the cost of a loan.
Quantifying credit losses helps in the decision-making—for instance, defining and monitoring the type of contract that should be promoted and rolled over, as well as the kind of counterparty linked to this contract, the limits for type and amount of borrower lending, etc. Moreover, some of the elements in credit and liquidity risk management and monitoring are based on the credit losses, e.g., credit losses impact directly the expected cash flows.We calculate expected credit loss based on two different assumptions: non-default events and default events. The former is based on the future expected exposures and default probabilities through the time intervals whereas the latter defines the actual estimated losses considering the expected collaterals and recoveries after the time of default event.
In the case of a non-default event, the consideration of time is defined up to T (i.e. maturity), or up to a given horizon date. The general formula of expected credit loss (ECL) for non-defaulted exposures is given by 9.4:
where
■ The sum covers all time iterations.
■ Pi is the discount factor from the current (analysis) date through the time interval (bucket) i to maturity / given horizon date. This factor includes the market risk free as well as any other discount spreads.
■ PDi is the probability of default, as discussed in Chapter 7, at the beginning3 of each time interval (bucket) i.
■ EEi is the expected credit net exposure of the contract(s), based on the exposure distribution at future times, as discussed above, at the beginning of each time interval (bucket) i.
In the case of default the expected credit loss is given by 9.5:
where
■ GE is the gross exposure, discussed earlier, at analysis date (with past due amount),
■ CE is the credit enhancements as discussed in Chapter 10 (Credit Enhancements), and
■ REC is the recovery amount as discussed in Chapter 8 (Behavior Risk).
The above equation 9.5 is also defined as the Loss Given Default and expressed as LGD = Enet - REC
As we have already seen in the estimation of credit losses there are several elements that need to be considered. Thus, ratings and default probability define both obligor's and guarantor's credit status. These together with the expected recovery will indicate the loss probability and severity. Moreover, all market risk factors, including credit speeds, and their correla- tions/volatilities need to be considered when asset-based credit enhancements are applied. Finally, as already mentioned, the value concepts will impact both current and future credit exposures and corresponding expected losses.
9.6