The past circa 15 months have been some of the most challenging within the industry and we have worked with many of our partners to help them successfully navigate the journey and remain in a strong and robust position. Businesses, and in particular SMEs, have changed – in part adapting to their circumstances but also from stimulus and intervention and continuing effects from restrictions. For lenders, how credit risk is assessed and measured has needed to change too and they have had to keep pace. There are growing reasons to be optimistic and, if approached correctly, financial institutions should be able to successfully navigate the remaining hurdles of Covid with regards to impairment and residual portfolio risk and move into a time of balance sheet growth in a robust and sustainable manner.

Background

With the continued roll out of Covid vaccines across the UK, there is a level of optimism that things may begin to return to a level of normality. Lockdown measures are being eased, albeit slower than hoped, and this is anticipated to lead to strong growth in economic activity. This growth or reigniting of the economy will require a level of further, likely lower risk, funding which presents an opportunity to lenders. However, balancing this opportunity with ongoing fallout from the last year, alongside unwinding decisions made during the pandemic, may materially impact the pace and approach taken by lenders in the future; Including the consideration of alternate decisioning algorithms and risk policies.

Willingness of banks to lend to SMEs had been improving post 2008 before falling back in the lead up to the pandemic period. A core development of lending into the UK SME market has been to open up what was traditionally a relatively closed market with limited providers and poor choices for customers. Lending was predominantly through the bank that held the current account. Growth has been driven through lenders access to a greater granularity of data meaning many decisions have been automated for lower value lending, reducing the cost to both the lender and business. Heading into the COVID period, understandably banks’ initial appetite for lending to businesses reduced significantly, due to potential impacts from the pandemic on their ability to trade as well as supply chains and distribution channels. The role of Governments were key to providing the backing needed to create stability, certainty and to expedite the flow of funding through to every business that needed it. Intervention supports – rebates, grants, deferred payments, loan schemes etc – have been comprehensive, and essential for maintaining a working nation. This has been alongside some private support for businesses such as negotiated cost reductions and rent deferrals from landlords.

SMEs account for c. 60% of employment in the private sector and therefore government intervention schemes such as the Coronavirus Business Interruption Loan Schemes were put in place to protect a large portion of the UK economy. Around 17% of SMEs had a demand for credit heading into the pandemic, of which nearly all was met due to HMT’s schemes, whereby over 90% of lending was through one of the Government backed loans. These facilities were designed to support any business, irrespective of size or industry, which had been impacted by the pandemic. The features of the schemes included guarantees to the lenders on losses, coverage of some interest payments and payment holidays at the outset (later extended to include the optionality for future payment holidays). The schemes increased credit supply for businesses most in need and resulted in north of 1.5m facilities being written. It is not yet clear precisely how many SMEs have been irrevocably impacted by the pandemic, but for now, given the low insolvency rates, at least the COVID facilities appear to have served their purpose.

Impact on Credit Decisioning

The credit outlook for businesses post-pandemic is uncertain. Many SMEs now have existing facilities to service and a reduction in historic turnover levels, therefore traditional affordability markers are likely to show reduced levels of affordability, but this may not reflect the underlying risk given many companies adapting to a low turnover environment and the flexibility of the government backed lending. Therefore, a lender’s ability to estimate credit risk is going to be impaired as a result of the pandemic and HMTs intervention meaning less reliance can be placed on traditional policies and models. It is envisaged that the purpose and usage of the credit taken now will become increasingly important with examples being seen of both SMEs who have a credit need for investment which have thrived in the pandemic and those businesses which have already taken a COVID facility and need another cash injection to stay afloat whilst footfall returns to the high street. The consequence of this reassessment of the ability for existing techniques to accurately reflect credit risk will impact throughout the asset lifecycle including onboarding, account management and risk measurement (impairment and capital).

Further to this already changed landscape, HMT have released the Recovery Loan Scheme (RLS) meaning there is still government guaranteed credit available to businesses which lenders might turn to. To be eligible for this scheme, facilities need to be underwritten as part of an RLS application, but the guarantee offers the lenders perhaps some certainty in their lending decision for borderline cases. Determining credit worthiness, at least in the near term, is going to be challenging for lenders through automated means with options being to accept without guarantee, accept with guarantee or reject. One option for lenders could be to revert to low-tech approaches through manually underwritten decisions. However, as has been seen for some time – low value, high volume lending, is unlikely to make economic sense to lenders.

We are now engaging with a range of lenders to develop new and advanced techniques including machine learning in decisioning to maintain high levels of automation, alongside enhanced returns models to account for greater uncertainty and ensure pricing is reflective of this. This work utilises our existing advanced analytical capabilities, gained through internal R&D investment and hands-on application, alongside our unique insights into the commercial lending industry during Covid and is based on key specialist engagements on large scale portfolios.

Future economic outlook

In addition to market changes, the credit and lending environment is against a backdrop of economic uncertainty – although the outlook compared to that at year-end has improved significantly. This improved outlook mainly reflects the successful vaccine roll-out, a modest immediate fall-out from the end of the Brexit transition period and the latest Budget which extended support to individuals and business, including the RLS already highlighted. However, all these aspects have not been fully resolved and therefore some caution is being seen with regards to the revised outlook.

There are, however, signs that the recovery will be robust as spending on debit and credit cards had returned to pre-crisis levels at the start of May and most households are happy with the state of their personal finances. This is being captured in many economic forecasts where renewed optimism features in several base case scenarios being used through financial organisations, including for their impairment reporting. However, whilst economists provide more optimistic macro-economic outlooks, these are not necessarily yet fully shared by risk professionals who, whilst acknowledging the potential for a strong recovery, note the impacts on many sectors, and in particular borrowers’ ability to repay, has not necessarily been fully realised in terms of defaults, repossessions etc. This, coupled with the fact that insolvencies have remained low throughout the pandemic, may highlight the positive impact of the government intervention but also potentially indicating a ‘bubble’ or undetected insolvencies that will be recognised when the support measures end.

Setting impairment levels

This level of unknowns against a backdrop of general economic optimism has required careful consideration with regards to financial reporting of balance sheet impairment and in particular the forward look requirement of IFRS 9. The narrative of scenarios varies across organisations, but can be reasonably expected to align to our own thoughts, being:

  • Baseline: It is assumed that vaccines remain effective against new variants of the virus and that households feel confident enough to resume much of ‘normal’ life. The measures introduced in the Budget in March prevent large-scale insolvencies.

  • Upside: In addition to the baseline assumptions, consumers spend freely, and economic output regains pre-crisis levels by the end of Q3 2021.  The post-Covid economy is very much like pre-Covid e.g., people return to the office. Business investment surges on the back of the super-deduction announced in the Budget.

  • Downside: The economy recovers, but the pace disappoints. This might reflect a reluctance among large parts of the population to emerge into crowded areas and a precautionary decision to hold onto savings.  The unemployment rate rises as the government decides that furlough is masking underlying problems rather than helping to solve them.

  • Severe: Vaccines prove ineffective against new variants. The economy might be locked down again and would contract early next year.  Supply-side constraints see inflation rise sharply. There is growing concern about capacity in haulage and construction industries given the loss of many EU workers.  At the same time, shipping costs are rising sharply as the global economy rebounds. This increase in inflation could cut into real incomes and exacerbate the recession.  House prices fall, reinforcing the fall in spending by reduced household wealth.

In terms of the impact of these scenarios on reported impairment levels, whilst we wait for the next financial updates, we can turn to the last banks’ financials and in particular the commentary of the changing economic outlook and how this has been translated into forward looking estimates for credit impairment. Firstly, when comparing scenarios, we saw that management statements acknowledged an improved economic outlook for the UK, mostly driven by a successful vaccine roll-out. Within these scenarios peak unemployment for most of the banks’ base case scenarios had dropped by c.1 percentage point for the Q1 2021 forecasts: resulting in peaks ranging from 6%-7%. However, in terms of application to financial accounts it is clear that initial model outputs are not being used without adjustment, reflecting the uncertainty stated earlier. This does cause an increased level of scrutiny from internal governance and auditors, as outputs from sophisticated models are being adjusted and overwritten based on little data or outcomes. This is not unsurprising and has been commonplace for over 12 months but, given the improving outlook, pressure is growing to introduce plans to unwind the overlays and return to a more business as usual approach. This is particularly as the basis for the adjustments are changing, from elements the model cannot be expected to account for – unprecedented economic forecasts, risk interventions (payment holidays, government schemes etc); to more general uncertainty that the worse is over.

In terms of methods to adjust their impairment outputs in recent reports, some lenders chose to retain the Q4 2020 forecasts as they favour a less optimistic view compared to Q1/Q2 2021 forecasts or have adversely updated economic weights in response to this improved outlook. As well as this more formulaic approach, there has been widespread usage of various post model adjustments (PMAs) to reflect economic uncertainty and to incorporate the impacts of ongoing government support suppressing expected default events. Whilst the use of expert credit judgement and PMAs have always been necessary and expected within the setting of impairment levels, the level of magnitude is significantly beyond that seen previously, particularly since the introduction of IFRS 9. Given the increased usage of these tools, the governance framework that oversees both the creation and approval of these values is also increasing to ensure full transparency, up to board level, of the rationale and approach for the setting of the value and expected roadmaps to unwind into the future.

With regards to PMA derivation, as the economic forecast has become more optimistic, the focus has been mainly on ensuring that the large impairment releases from updated scenarios is managed to reduce volatility. This has been achieved through the inclusion of new PMAs to account for uncertainty in the new forecasts or the retaining of historic PMAs (where seen to increase impairment).

Historic PMAs through the Covid pandemic took many forms with some lenders smoothing their forward-looking forecast (final output) around the COVID data points e.g., averaging the sharp fall in GDP in 2020 Q2 followed by a rebound, whilst others used a longer-term projection of 3–4 years, with limited focus being placed on the volatile 2020 period with the expectation of it being a temporary one-off event with limited long-term impact. In addition, judgement-based approaches have been widely used including using turnover/income shock data to identify at risk business, assessing payment holiday distributions to assess accounts most susceptible to default and sector specific data to allow overlays for riskier sectors such as hospitality.

Summary

Overall, the past 15 months have been some of the most challenging within the credit industry and we have worked with many of our partners to help successfully navigate the journey and remain in a strong robust position, through careful impairment management, more efficient collections processes and enhanced reporting. Focus now turns to the work required in preparation to seize the growth opportunities to come.

There are now growing reasons to be optimistic, and a further quarter of improved economic outlook will likely see further release of impairments and greater balance sheet resilience. This in turn should lead to greater comfort with increased lending, although the pricing of individual exposures and the decision of whether to use the new RLS guarantee scheme will still cause challenges – particularly when relying on automated models, seen as a necessity for economical high volume, low value lending.

We believe that, approached correctly, financial institutions should be able to successfully navigate the remaining hurdles of Covid with regards to impairment and residual portfolio risk and move into a time of balance sheet growth in a robust and sustainable manner.

If you want to speak to 4most about how we can support with credit risk within commercial lending, please contact Chris at christopher.warhurst@4-most.co.uk