13.1

LFIs must have methodologies and analytical solutions to measure, analyse and categorise Credit Risk, and compute the associated provisions. LFIs must be able to analyse Credit Risk at several granularity levels including Credit Facility level, Obligor level, segment level and portfolio level in order to identify credit concentration risk.
 

13.2

Each LFI must operate analytical tools with sophistication appropriate to the complexity of its portfolio, products, industries and other predominant factors. If an LFI does not use models for a given portfolio, then it must document and provide a rationale for such a decision.
 

13.3

Where models are used for decision-making, LFIs must articulate clearly the roles of models for the support of Credit Risk underwriting, monitoring and provisioning. They must define the modelling strategies, the limits and conditions of model usage. Particular attention must be given to any overriding of model outputs. Such practice must be justified, fully documented and reported. High frequency of overrides of model outputs should be remedied by model recalibration or development.
 

13.4

LFIs must establish an appropriate asset grading or classification system for the measurement of Credit Risk. All acquired and existing Credit Facilities must be assigned a rating grade based on robust justification and supported by historical analysis and analytical tools.
 

13.5

When an LFI makes use of models for decision-making for credit purposes, it must also comply with the modelling standards and Guidance issued by CBUAE. In addition, LFIs must ensure the establishment of effective controls (including in respect of the quality, reliability and relevance of data and in respect of validation procedures) around the use of models to identify and measure Credit Risk and set limits. Each LFI must demonstrate that its models are fit for purpose and adequately calibrated to effectively support the associated risk and business decisions.