Credit Risk Modelling & Provisioning for Banks: Managing the potential impact of COVID-19

A Southeast Asia perspective

Implications of covid-19 on provisioning

Post the global financial crisis (GFC), regulators worldwide formulated plans which included requirements that financial institutions must hold provisions which make allowance for expected economic impacts.

Banks across the globe have been working towards developing sophisticated forward-looking provisioning models to allow for different simulated economic scenarios. These simulations include “downturn” economic scenarios which are designed to stress test the bank’s provisions and capital levels.

However, the COVID-19 pandemic today has brought many economic activities to a virtual halt, along with muted GDP growth worldwide, as forecasted by various public and private organisations. 

Understanding the impact on financial reporting

One of the major challenges for banks during this period would be to meet the key requirement under the International Financial Reporting Standard (IFRS 9), i.e. a forward looking approach which recognises impairment losses in a more timely manner.

The Standard was designed in view of the weaknesses identified in the previous reporting standards and delayed recognition of losses during the GFC. The global economy picked up and has been relatively stable since. 

However, with the current COVID-19 pandemic, it will be a challenge for most banks across the world that are reporting under IFRS 9 to meet the key requirements of the Standard in computing “forward-looking” expected credit losses (ECLs). As the IFRS 9 Standard requires banking institutions to incorporate “probability-weighted scenarios” when computing ECLs, the credit modelling teams within banks will need to re-think how forecasted economic shocks and respective probability weights can be incorporated into existing impairment models.

In times of economic stress,  here are four common adverse effects seen around Southeast Asia:

Increased Loss of Income

Increased Loss of Income

Retail borrowers and commercial borrowers are facing significant reductions in their monthly incomes.

Plunge in Economic Activities

Plunge in Economic Activities

Lockdowns are leading closure of several business and disrupting supply chain and mobility.

Increased Risk of Default

Increased Risk of Default

During this outbreak, many lenders are facing breach of loan contracts by their borrowers.

Low Financial & Operational Leverage

Low Financial & Operational Leverage

Slowdown in economic activities is putting pressure on businesses’ cash flows and working capital.

Relief measures are deployed across Southeast Asia to minimise the impact due to economic shocks

Here is an overview of regulatory moratoriums in Singapore, Malaysia, Thailand, Philippines, Indonesia and Vietnam:

Monetary Authority of Singapore (MAS)

  • Letting borrowers with secured loans defer either principal payments or principal & interest payments until the end of the year as long as the borrowers are not in arrears for more than 90 days
  • No interest charged on the deferred interest payments
  • Flexibility to convert unsecured loans to term loans with a reduced rate of interest. This option is specially designed to support borrowers who suffered a huge loss of monthly income (more than 25%)
  • Deferring principal payments on SME secured loans until the end of the year
  • Potentially lowering interest rates on some of the SME credit facilities

Bank Negara Malaysia (BNM)

  • BNM’s measures announced that the banks will let both retail and commercial borrowers to defer repayments for a period of 6 months
  • Some of the major Malaysian banks also mentioned they would not compound any interest on the deferred payments for this period
  • Banks will also allow individuals to convert some of the outstanding balances of unsecured credit facilities to a term loan for not more than 3 years
  • Banks will also now assist corporates in restructuring some of their loans to help with sustainability of businesses and jobs

Bank of Thailand (BOT)

For customers who are not in arrears for more than 90 days, have not gone through debt-restructuring and not undergoing legal proceedings:

  • A grace period of up to 1-year on the repayment of the principal
  • A reduction of the instalment amount of up to 1 year
  • A six-month maximum grace period for instalment payments

Bangko Sentral Ng Pilipinas (BSP)

Major banking institutions and banks:

  • To exclude from the past due loan ratio of loans to affected borrowers for a period of one (1) year
  • Stagger booking of provision for probable losses over a period of five (5) years, subject to prior approval of the BSP
  • Flexibility to provide a grace period of 30-90 days for loan payments and/or restructure loan accounts for their customers
  • To relax eligibility criteria for borrowers affected by the COVID-19 pandemic
  • Strongly encouraged to suspend fees and on-top charges for online banking transactions to facilitate smooth banking experience for customers

Otoritas Jasa Keuangan (OJK)

  • Banks to relax debt quality assessment and restructuring requirement for affected debtors (including individuals and micro, small and medium enterprises (MSMEs) for a period of up to 12 months
  • Banks encouraged to ease up on debt collection process for affected individuals who are working in informal sectors or workers with daily income

State Bank of Vietnam (SBV)

  • Banks to reconsider debt repayment terms, exempting/lowering interest rates in view of the customers who suffered great loss of income and decrease in revenues as a result of COVID-19
  • Financial institutions advised to extend credit lines to businesses and exempt or reduce interest payments. Some businesses are given flexibility to extend the loan maturity to a maximum of 12 months from contractual maturity

Key areas where banks can focus their efforts

Banks to adapt to the current situation and incorporate government moratoriums

IFRS 9 requires most financial institutions to determine forward-looking expected credit losses. However, due to major economies coming to a near shutdown and uncertainty creeping into economic activities, banks are finding it extremely challenging to quantify the impact with the existing forward-looking lenses or models in place. Additionally, banks may experience a significant increase in ECLs which would mean tightening some of the existing credit policies.

Regulators and governments are mandating moratoriums such as payment holidays etc. This would require banks’ credit modellers to take note of these moratoriums and ensure some of these loans (under payment holidays policies) are not transferred into ‘Stage 2’ / ‘Stage 3’ classifications (attracting lifetime ECL) purely based on the 30 days past due criteria.

Banks are now expecting an increase in IFRS 9 provisions which would also have a significant impact to the banks’ capital requirements. At present, major banks in Southeast Asia appear to have enough capital cushion, but the uncertainty in how long the pandemic will last is cause for concern amongst the banking institutions. Key areas that banks should focus on given the current turbulent situation are:

  • Economic scenarios
  • Staging Criterias
  • Underlying Credit Models

Banks to start capturing reliable and robust economic scenarios for forward-looking provisioning

Many central banks acknowledge the fact that due to sudden and unexpected effects of COVID-19, it is difficult to obtain reliable and detailed macroeconomic forecasts to calculate forward-looking ECLs. Under these circumstances, it will be important for banks to rely on robust qualitative overlays and expert judgments applied on the model outputs to determine IFRS 9 provisions.

For example, for retail banking portfolios, many banks, especially in the Southeast Asian region, can expect macroeconomic variables like household debt, income, unemployment rates, and interest rates etc. to drive some of the risk factors. As most macroeconomic forecasts are concerned, the above-mentioned variables may generally be offset and computed based on the performance of GDP. However, during this outbreak, a large fall in GDP may not necessarily mean an increase in unemployment rates but may mean a decrease in retail expenditure. This could be due to companies looking to retain employees, but at lower remuneration levels. Banks’ management teams will have to closely monitor these variables and apply
post model adjustments to capture the current challenging conditions.

Flexibility in applying Staging triggers based on regulatory moratoriums

Most banks should expect that many of their existing loans will be triggered into “Stage 2” from “Stage 1” given the current economic conditions, which also means a significant increase in provisions. However, for most banks, it will be arduous to source reasonable and supportable information at the reporting date at individual or loan level when assessing increase in credit risk.

Banks may consider incorporating additional macroeconomic information at the reporting date to perform the Staging assessment and potentially consider further segmentation of their existing portfolios to identify vulnerable pools/segments that may need to be placed under “Stage 2” bucket.

In the short-term, the management teams in the bank may have to apply factors based on the moratoriums released by the regulators to ensure there is no sudden spike or overestimation in provisions due to loans transitioning to “Stage 2” or “Stage 3” during this period. This is also likely to have an impact in the way the bank captures, tracks and reports these customers over time. It is highly recommended that banks work with and seek guidance from their regulators to finalise their first quarter provisions.

Banks’ internal credit risk teams should continue working on monitoring and enhancing risk models

The banks’ internal credit risk management teams should continue to work on enhancing the existing credit models and allow for the measurement of credit risk with and without any moratoriums. Banks may also consider recognising significant increase in credit risk, under IFRS 9 requirements, without making allowance for the impact of the pandemic. This could then be allowed for in an additional overlay.

Existing forward-looking or macroeconomic linkage models that help predict default risk should capture those variables that have been directly affected by COVID-19. Variables considered as predictors may no longer be relevant. Structural breaks in economic variables and default / loss rates should be tested for when developing and enhancing forecasting models.

Many banks in Southeast Asia have been facing challenges from a data collection perspective :

  • Banks will need to continue collecting data points at requisite levels of granularity to ensure for robust modelling and stable model outputs
  • Capturing reliable macroeconomic data and forecasts will be very critical during this outbreak

Helping you navigate through these challenging times

Banks’ internal risk processes, policies, models and systems are very much in the eye of the storm right now. As retail and commercial customers come under financial stress for cash flow and loan repayments, banks have difficult decisions to make about how to help clients while maintaining their own balance sheets. IFRS 9 has onerous modelling and provision implications which will not be sustainable in the current environment. Governments are stepping in with relief programmes to help struggling consumers and businesses.

It is important to consider substantial correlation of various other risks, including but not limited to credit risk, market risk, operational risk, liquidity risk, cyber security risk etc. Linkages between different categories of risk are likely to emerge in times like this and should be fully understood and prepared for. Most banks will be under scrutiny during this period and will be closely watched on how the internal risk systems and models cope with the current turbulent environment.


We understand this period will be a challenge for many banks and we welcome discussions in navigating approaches and considerations in enhancing internal risk models based on the ever-changing regulatory requirements.

Follow us

Contact us

Antonie Jagga

Antonie Jagga

Partner, PwC Singapore

Tel: +65 9667 5825

Dickson Wong

Dickson Wong

Director, South East Asia Consulting, PwC Singapore

Tel: +65 8299 9721

Hide