CA-5.8.57

This section addresses the broad standards for own-estimates of PD, LGD, and EAD. Generally, all banks using the IRB approaches must estimate a PD⁵⁰ for each internal borrower grade for corporate, sovereign and bank exposures or for each pool in the case of retail exposures.

⁵⁰ Banks are not required to produce their own estimates of PD for certain equity exposures and certain exposures that fall within the SL sub-classes.

CA-5.8.58

PD estimates must be a long-run average of one-year default rates for borrowers in the grade, with the exception of retail exposures (see below). Requirements specific to PD estimation are provided in paragraphs CA-5.8.72 to CA-5.8.78. For retail asset class banks must estimate an appropriate LGD (as defined in paragraphs CA-5.8.79 to CA-5.8.83) for each of its retail pools and must also estimate an appropriate long- run default-weighted average EAD for each of its facilities as defined in paragraphs CA-5.8.84 and CA-5.8.85. Requirements specific to EAD estimation appear in paragraphs CA-5.8.84 to CA-5.8.89.

CA-5.8.59

Internal estimates of PD, LGD, and EAD must incorporate all relevant, material and available data, information and methods. A bank may utilise internal data and data from external sources (including pooled data). Where internal or external data is used, the bank must demonstrate that its estimates are representative of long run experience.

CA-5.8.60

Estimates must be grounded in historical experience and empirical evidence, and not based purely on subjective or judgmental considerations. Any changes in lending practice or the process for pursuing recoveries over the observation period must be taken into account. A bank's estimates must promptly reflect the implications of technical advances and new data and other information, as it becomes available. Banks must review their estimates on a yearly basis or more frequently.

CA-5.8.61

The population of exposures represented in the data used for estimation, and lending standards in use when the data were generated, and other relevant characteristics should be closely matched to or at least comparable with those of the bank's exposures and standards. The bank must also demonstrate that economic or market conditions that underlie the data are relevant to current and foreseeable conditions. For estimates of LGD and EAD, banks must take into account paragraphs CA-5.8.79 to CA-5.8.89. The number of exposures in the sample and the data period used for quantification must be sufficient to provide the bank with confidence in the accuracy and robustness of its estimates. The estimation technique must perform well in out-of-sample tests.

CA-5.8.62

In general, estimates of PDs, LGDs, and EADs are likely to involve unpredictable errors. In order to avoid over-optimism, a bank must add to its estimates a margin of conservatism that is related to the likely range of errors. Where methods and data are less satisfactory and the likely range of errors is larger, the margin of conservatism must be larger. CBB may allow some flexibility in application of the required standards for data that are collected prior to the date of implementation of this Module. However, banks must demonstrate to CBB that appropriate adjustments have been made to achieve broad equivalence to the data without such flexibility. Data collected beyond the date of implementation must conform to the minimum standards unless otherwise stated.