IFRS 9 modelling: Jaywing extends use of Horizon platform to incorporate static models
Upcoming AI regulations and how to get ahead of them
PRA Collections Audit - what lenders need to do now
20 May 2021
The Prudential Regulation Authority’s findings from their internal audit review of non-systemic banks and building societies were issued in a letter to Chief Risk Officers last month. In this article, Neil Roberton, Data Science Director, reflects on the findings, and discusses what Collections teams need to do next.
The PRA’s review commenced in November 2019 and was primarily concerned with the effectiveness of controls in three areas:
Collection processes and control environment
Governance and oversight
It was commissioned as a result of concerns about the effectiveness of organisations’ Collections capabilities and their ability to ramp-up efforts under a stressed environment – a scenario that could very soon become a reality given the withdrawal of Covid-19 support programmes, such as the furlough scheme.
42 firms were reviewed, and the findings offer an interesting insight into the current challenges that organisations face, and the areas of focus for the PRA. Whilst the overall message was positive (none of the selected firms were given a Red rating), there were some clear notes of caution, with 20% of firms receiving an Amber rating overall.
The primary areas of concern for the PRA were the ability of firms to handle forbearance processes and associated reporting, and the lack of control over Collections strategy execution. These are both flagged as being critical: it is only through having a tight grip on Collections strategies and processes during this period of relative benign uncertainty that firms will be able to respond rapidly and accurately when it comes to an end.
Timing is crucial
Now is the time for Collections functions to rise to this challenge, both to satisfy the regulator and to achieve the best outcomes for their customers and stakeholders. Whilst overdue volumes and impairment numbers are lower than anticipated in this year’s budgets, there is an inherent lack of certainty around the two core drivers of non-performing debt as a result of the pandemic:
Discretionary spending has been reduced. During this year’s lockdown, spending on holidays, eating out and clothes has been reduced. Typically, it is this type of expenditure that can overstretch consumers’ ability to repay debt. The gradual loosening of restrictions is likely to result in a “catch up” spending release, as people return to discretionary spend. This will result in people who were consistently managing and paying down their balances taking on increased debts that they are subsequently unable to afford.
The withdrawal of income protection measures will start to bite. Measures to support businesses and individuals during the pandemic protected up to 80% of people’s income. This support was particularly important at the lower end of the income scale. As this support reduces, it is likely that many incomes will not return to pre-pandemic levels. This could be another critical factor in determining whether people are able to continue to meet their debt repayments.
How firms can respond
The PRA letter is clear on how firms should prepare for the challenges ahead: minimise losses, but ensure customers are treated in a fair and consistent manner, taking account of individual circumstances. This is critical, and offers opportunities to develop positive, longer-term customer relationships. But there is a risk of reputational and regulatory damage where firms do not reach the standards expected.
Firms must ensure they have clear Collections processes that are understood, documented and controlled. In order to meet the bar that the PRA has set around controls and quality assurance, firms should consider these points:
Understand as much about each customer as possible. Use all available data to identify individuals that are showing early signs of financial distress. Include internal and external performance information, as well as transaction histories and income information where possible, and use predictive models to drive accurate identification.
Ensure that this data is up-to-date and accurate. The data used to select customers for collections activities must be highly accurate. Apply automated data quality checks to spot potential errors before they result in inaccurate customer interactions - preventing confusion and avoiding misleading or reputationally-damaging communications.
Deploy transparent segmentation and strategy allocation to increase response. Clearly document the processes for allocating customers to contact groups based on likelihood to self-cure. Establish clear, transparent segments to distinguish between high and lower-risk cases at the pre-delinquency stage. Ensure that customers receive a progressive, supportive collections journey avoiding sporadic or too little contact which can reduce the likelihood of payment from a financially distressed consumer. Seek to understand underlying non-payment reasons and strive for a mutually beneficial conclusion.
Develop clear Management Information. The PRA is specific in its requirement for timely and accurate reporting, particularly around the impact of forbearance measures on customers and on firms’ balance sheets. Understand the metrics and processes behind forbearance measures, ensuring accurate tracking. By doing this, you can demonstrate that appropriate measures are available where necessary, closely controlled and monitored to prevent deeper repayment problems being stored for the future.
Continuous, controlled improvement. Implement and continually refine a transparent, controlled, test and learn framework. Test different contact strategies for different Collections segments and channels, to identify improvements and understand the borderline between different groups. This is a critical element of capacity planning, optimising the use of scarce operational resources despite increasing collections volumes. Digitise where possible using self-serve to minimise operational cost.
5 steps to transform your collections processes:
Use the latest AI approaches to develop smart, highly accurate models
Use data from a variety of internal and external sources to better customer behaviour and the potential reasons for non-payment
Use a continuous improvement approach, to ensure that you know which strategies, channels and tone are working best for each group of collections customers.
Ensure that MI can be easily understood and provides a clear understanding of operational & strategic performance as well as identifying areas that can be adjusted to account for a Covid-driven spike in collections volumes.
Ensure that operational capacity plans are closely aligned to the objectives and forecast of the wider business e.g. what are Risk seeing regarding the underlying quality of the portfolio and how is this changing over time.
Our experience in Collections
Since 2001, Jaywing has worked with providers across the financial services sector to pioneer and perfect the use of data within their collections functions. We have supported a number of clients with their collections strategies and processes, both in lending and debt purchase. This support has ranged from high-level strategic reviews and on-going consultancy, through to bolstering client’s teams with hands-on analytical and data developer support, as well as digital Collections solutions. We have also supported clients with their journey towards purchasing a decision system, including support for building a business case to justify this investment. Our modelling software guarantees highly predictive Collections models using your historical data. In short, we are well placed to help you to meet your requirements under the PRA review.
For more information, please get in touch.