To: Locally incorporated authorised deposit-taking institutions (ADIs)

Observations on provisioning practices

1. Controls around model risk management

• Many ADIs continue to apply sizeable judgment-based adjustments to compensate for model and data limitations, to ensure provisions reflect credit risk expectations. The focus of these ADIs has shifted towards adjustments to capture the impact on borrowers of rising interest rates, high inflation, and commodity market volatility. 
  

• APRA acknowledges that judgement-based adjustments have a role to play, and models alone may not be calibrated to adequately capture and reflect all drivers of credit risk. Such adjustments include the addition or removal of overlays, based on expert judgement, where appropriate, to ensure that provisions reflect actual credit risk expectations. Where judgmental adjustments are made, APRA expects that they be supported by appropriately documented analysis and robust controls and governance. This includes sound processes for the application of adjustments, controls to ensure the efficacy of the output, and senior management oversight and accountabilities. 
  

• While APRA recognises the need for judgmental adjustments, it is important that ADIs monitor and continuously improve model performance for credit quality assessments and provisioning.

2. Capturing economic uncertainty

• Sensitivity analysis is used by ADIs to capture the impact of macro-economic headwinds on borrowers. ADIs often use sensitivity analysis to understand how provisions respond to changes in the key underlying drivers of credit risk. 
  

• In the current environment, it is important that ADIs can perform comprehensive sensitivity analysis on a regular and timely basis across credit portfolios (by appropriate industry, geographical and other segmentations). This can support management and Boards in understanding the sensitivity of the credit portfolio to key credit risk drivers and help to inform appropriate provisioning levels. Such sensitivity analysis could form part of an ADI’s overall stress testing capability, as part of the ICAAP and risk appetite review processes.[3]
  

• As with judgmental adjustments, APRA expects that appropriate governance to be in place for sensitivity analysis, including clear accountabilities, robust controls and oversight. This underpins both the process of, and decisions made, in the sensitivity analysis exercise.

3. Identifying credit deterioration in vulnerable sectors and borrowers

• Many ADIs continue to apply manually intensive, and judgement driven processes to understand and capture the impact of emerging risks on vulnerable borrowers and sectors.  However, it is important for ADIs to invest in systematic processes to identify vulnerable sectors and factor sectoral risks into loss estimates. These processes would include the use of collective assessments, the level of segmentation of models and data, and the indicators used to transfer loans in vulnerable sectors into impairment Stage 2 under Accounting Standard AASB 9 Financial Instruments.