5.2.4 Credit risk measurement

Credit risk measurement

Loans and advances (including loan commitments, LCs and LGs)

The estimation of credit exposure for risk management purposes is complex and requires the use of models, as the exposure varies with changes in market conditions, expected cash flows and the passage of time. The assessment of credit risk of a portfolio of assets entails further estimations as to the likelihood of defaults occurring, of the associated loss ratios and of default correlations between counterparties. The Group measures credit risk using Probability of Default (PD), Exposure at Default (EAD) and Loss Given Default (LGD). This is similar to the approach used for the purposes of measuring Expected Credit Loss (ECL) under IFRS 9.

Credit risk grading

The Group uses internal credit risk grading that reflect its assessment of the probability of default of individual counterparties. The Group uses specific internal risk rating models tailored to the various categories of industry/segments of counterparty. Borrower and loan specific information collected at the time of application (such as financial spread, management quality, Risk Bureau information, and account conduct, turnover and industry type for wholesale exposures) is fed into this rating model. This is supplemented with external data input such as credit bureau scoring on individual borrowers into the model. In addition, the models enable expert judgement from the Credit Risk Officer to be fed into the final internal credit rating for each exposure. This allows for considerations, which may not be captured as part of the other data inputs into the model.

The credit grades are calibrated such that risk of default increases exponentially at each higher risk grade. For example, this means that the difference in the PD between a 6 and 8 rating grade is lower than the difference in the PD between an 18 and 20 rating grade.

The Risk Rating system for performing assets ranges from Obligor Risk Rating (“ORR”) ORR 1 to ORR 7, each grade being associated with a Probability of Default (“PD”). Non-performing clients are rated ORR 8 to ORR 10, corresponding to substandard, doubtful and loss classifications.

Retail

After the date of initial recognition, for retail business, the payment behaviour of the borrower is monitored on a periodic basis to develop a behavioural score. Any other known information about the borrower which impacts their creditworthiness – such as unemployment and previous delinquency history – is also incorporated into the behavioural score. This score is mapped to a PD.

Retail risk parameters i.e. PD, LGD & CCF or EAD have been estimated using ‘Segmentation Methodology’ or ‘Retail Pooling’; where following factors have been considered:

• Asset classification as Credit Card, Executive Finance (or Personal Loan) and SME (or small business loan);
• Days Past Due – 2 segment each for Stage 1 & Stage 2; and 1 for Stage 3

Risk parameters have been estimated with respect to above segments and used the same for ECL computation.

Wholesale

For wholesale business, the rating is determined at the borrower level. A relationship manager will incorporate any updated or new information/credit assessments into the credit system on an ongoing basis. In addition, the relationship manager will also update information about the creditworthiness of the borrower every year from sources such as financial statements. This will determine the updated internal credit rating and the mapped PD.

Treasury

For debt securities in the Treasury portfolio, external rating agency credit grades are used. These published grades are continuously monitored and updated. The PD’s associated with each grade are determined based on realised default rates over the prior 12 months, as published by the rating agency.