Qualitative Adjustments are Key to Modifying CECL Estimates for COVID-19 Impacts
How is it possible to estimate expected credit losses over the contractual life of financial assets in an environment that drastically changes day-by-day? In the three months since CECL became effective, the economic outlook has rapidly deteriorated as a result of the fallout from COVID-19. Because this is completely unchartered territory for many, how do you go about adjusting to this completely uncertain and volatile environment?
Previously, we discussed how some significant CECL assumptions have likely changed as a result of the pandemic. Two examples are the representative period of historical losses used for the foundation of the CECL estimate, and the forecasts over the reasonable and supportable period. It is much easier to discuss adjusting the CECL model to incorporate these changing assumptions than it is to make last-minute modifications within the model. Because the model has already been validated and reviewed by internal and external auditors, most filers find that it’s too late to adjust it for 1Q2020 CECL estimates and maintain robust model governance and control. DFAST and CCAR models are already designed to capture these more extreme and unexpected impacts in the severe scenarios, further reducing the need for CECL model changes.
Significance of Qualitative Adjustments
Remember those old qualitative adjustments that some thought would disappear with CECL? They never really went away, and have become even more important with the everchanging economic landscape wrought by the COVID-19 pandemic. Qualitative adjustments are used to account for uncertainty and limitations in the process based on management’s experience and judgment, and many institutions are using them to incorporate changing assumptions in their CECL estimates for first quarter.
Historical Credit Losses
First, consider the period and calculations used for the historical credit losses. An entity may only have data for more stable positive economic outlooks, such as the period after the financial crisis of 2008. In these situations, an entity could purchase industry data with credit loss experience over a broader economic environment or they could use their own, but overlay a qualitative adjustment on the model output to accommodate for the more severe current economic conditions. In these periods of extreme and unexpected volatility, the model results will likely need to be adjusted in order to consider a more adverse set of historical credit losses for use in projecting the CECL estimate.
Reasonable and Supportable Forecast Periods
Has your reasonable and supportable forecast period changed? The current situation is extremely volatile, and the model assumptions shouldn’t be changing every period based on instability alone. The reasonable and supportable forecast period chosen for use in the CECL estimates was based on the general period over which credit losses and related assumptions can reasonably be estimated. We continue to see industry players adjust for this by layering in on-top qualitative adjustments to the model results.
Directly changing model inputs and assumptions each period based on market volatility while also maintaining controls and governance over the modeling environment is very difficult. That’s where qualitative adjustments come in handy. Their magnitude is subjective, but the direction of change and comparison to previous CECL runs should provide valuable insight to management and consumers of financial statements. Disclosures and documentation are even more important now under CECL in this unpredictable market environment.
Our latest guidebook provides leaders with a roadmap to enhance resiliency plans, simplify operations, address new financial requirements, and more. To download, please click the link below.