The Data Analytics Blog

Our news and views relating to Data Analytics, Big Data, Machine Learning, and the world of Credit.

All Posts

10 ways the COVID-19 crisis will affect your credit models (PART 2)

April 8, 2020 at 7:11 PM

This is the second of a 2-part blog. You can read the first blog here.

At Principa we live and breathe models. In this blog we list 10 key ways how we believe COVID-19 will affect your credit models. It’s in 2 parts (this is the second part). The 10 items are as follows: 

PART 1:

  1. Bureau scores
  2. Application scores
  3. Acquisitions scores
  4. Behavioural scores
  5. Collection scores

PART 2:

  1. Payment holidays
  2. IFRS9
  3. Sales and budgeting models
  4. SMEs
  5. Vintages

In a future blog we will cover some remedial actions and dive deeper into some of these topics. 

We’ll be discussing the contents of the blog in a Webinar. To sign up email us here.

If you would like to chat to us to understand how we might be able to help you with your pressing concerns, please contact us here.

  • Payment holidays

Many credit providers, particularly banks, are already offering payment holidays to stressed consumers. What does this mean and how does this affect the health of a credit book where one has granted a payment holiday?  For some accounts you will be issuing welcome short-term relief for a short-term hiatus. For many, however, you will be delaying the inevitable as COVID-19 will (best case scenario) reduce SA’s GDP growth by 4.2% to 6% with a resulting loss of thousands of jobs. This means that provisions for those months of payment holiday will be understated and the behavioural scores (and PDs) will be higher than they would otherwise be.  With payment holidays it will be essential to be conservative with any expert adjustments to provisions and limit any aggressive extension of credit to the mid-range risk tranche.

  • IFRS9 models

We expect some significant impact on IFRS9 provision models.  Some initial analysis and scenario forecasting that  we have done has indicated a doubling of provisions in the consumer sector.  We expect this to be even higher on small business books. In summary some of the impacts of IFRS9 models include:

  • Payment holidays may warp PDs (models need to be able to exclude this period)
  • Markov models will initially understate flows, and if the economy eventually picks-up may over- state flow-rates.  Period exclusions may be required – or a matrix freeze. 
  • Run-off triangles will also ultimately be affected through recovery transitions.
  • Economic forecasting may require more extreme scenarios, but some economic indicators (e.g. stock exchange stability, job losses, $ exchange rate) may already represent some of these extreme scenarios. Scenario testing should avoid double counting.
  • Payment holiday relief and Solidarity and Billionaire Club relief funds may soften the blow on a temporary basis but may delay the inevitable (especially around SMEs) in the long run.  The new reality post-COVID-19 could return to “normal”, could have a “lasting impact” where credit performance and roll-rates are deteriorated on a permanent basis (e.g. due to increased unemployment, revenue losses, SME closures etc) or it could provide an “transformational opportunity” where the credit metrics settle at an improved level post-COVID19 (possible due to new online, bot, debit order and remote collections capabilities).
  • A key question that would need to be established: what is the new-normal with respect to PD, LGDs and EADs?
  • Once the new-normal with respect to PDs, EADs, LGDs, default timing, cure rates, economic relationships etc. is determined and the data have matured, models would require a refresh of back-testing to validate the continued use and subsequently might require alignment or rebuild.
  • Consider updating scorecard model monitoring packs to include “very” short term outcome windows to confirm stability, accuracy and predictive power over 1, 3, 6, and 12 months outcomes to ensure early warning triggers when models move away from theoretical expectation more than acceptable threshold (see point #10), this might be where ECL provisions are maintained on an expert basis (e.g. stable coverage ratios for a defined period of time).
  • You would need to exclude 2020 from any future modelling dataset.

We’ll unpack this a bit more in a future blog. Subscribe at the bottom of this page to receive notification.

  • Reduction in sales

For those offering revolving credit (but particularly retailers), the reduction in sales will mean the account profitability picture will look quite different. Over the lockdown period significant reduced spend is forecast which will affect profit forecasting and budget models.  It is likely to make some tranches of the portfolio unprofitable.  This is likely to be a temporary effect most harshly felt during the lockdown period, but how these accounts perform post-lockdown will be critical.

  • SMEs

SMEs are being heavily hit by the COVID-19 lockdown.  Whilst there’s good news with the Solidarity Fund and funding from Johan Rupert, the Oppenheimers, Patrice Motsepe and Naspers (amongst others) helping keep businesses afloat in the short term, it is still estimated than thousands of business will go under with a loss of thousands of jobs.  So, what are we likely to see? Whilst credit scores can be a good indication of companies under stress, this crisis has meant that many previously stable companies are under severe stress with reliable revenue streams drying up and the burden of wage bills, rent and trade-credit debt remaining. The apparent lowest risk tranche determined by a credit score will now see numbers of new delinquencies.  Scorecards are likely to reduce in risk discrimination strength as a result. Lenders will be forced to tighten up extension of credit across all scorebands and to monitor their book carefully. Lenders would also need to best understand which industries (for example tourism and hospitality) are being affected most and potentially have separate treatments for businesses in these industries. We expect provisions to escalate over the next 6-12 months.

  • Vintages

    We have spoken about the importance of monitoring, but it’s worth looking at some of the reports than should be potentially edited.  One of the key credit committee reports is monthly/quarterly vintages.  Here the default rate of accounts from different application months are monitored through typically 3-month aging stages.  These graphs give an early indicator of whether accounts from a certain application period are performing significantly worse than other periods.  During the COVID-19 lockdown period and aftermath, it may be worth producing vintages that look at early delinquency (e.g. CD1/CD2) to allow the business to take early action.

    Other worthwhile reports would be to again change the performance definition, but to look at score-to-odds relationship (for example CD1-ever after 3 months).  This would give an overall indication of how badly the scorecards have been affected.

    BI tools will also be essential to really understand what parts of the book are most adversely affected by the crisis. 

    Conclusion

    We expect the impact of the COVID-19 crisis on credit models to be widespread and quite significant. This black-swan will not be leaving us any time soon.  We would strongly recommend as a lender you consider addressing these effects by assessing them and taking remedial action.  If you would like to chat to us to understand how we might be able to help you with your pressing concerns, please contact us here.

    Contact Us to Discuss Your data analytics Business Requirements

     

Thomas Maydon
Thomas Maydon
Thomas Maydon is the Head of Credit Solutions at Principa. With over 17 years of experience in the Southern African, West African and Middle Eastern retail credit markets, Tom has primarily been involved in consulting, analytics, credit bureau and predictive modelling services. He has experience in all aspects of the credit life cycle (in multiple industries) including intelligent prospecting, originations, strategy simulation, affordability analysis, behavioural modelling, pricing analysis, collections processes, and provisions (including Basel II) and profitability calculations.

Latest Posts

Why Principa’s FinSmart is superior CreditTech

Principa’s FinSmart has become the industry’s go-to solution set for end-to-end credit management. Our credit risk management software products reduce risk and improve profitability by streamlining processes, increasing efficiency, and automating data-driven decision making across the credit lifecycle.

How chat is revolutionising the digital onboarding experience

Principa’s onboarding chatbot solution; Atura allows lenders to engage a customer effectively through an application process while accessing necessary data and decisioning calls using Principa’s SmartSuite software. The digital revolution “Digital” has been a financial services buzz-word for some time. Most South African lenders Principa works with have been working hard to adapt to a digital existence for several years. Some have been successful, others are still working on the challenge - and most have only partially adapted.

How to choose the correct collections chatbot

Principa has a wealth of experience in building and deploying chatbots for the financial services industry. Our custom-built solution is flexible and fully customisable which allows your bot to assume your brand’s persona. We can also seamlessly integrate with existing systems. Click here to find out more.