Managing the Complexity of CECL, Variances in Forecasted Credit Losses
The Current Expected Credit Loss (“CECL”) accounting standard requires entities to forecast credit losses expected over the life of a pool of loans. Estimating lifetime losses is complex and requires significant analysis. Regardless, at some point during the life of a loan pool, virtually every modeled forecast result will likely experience a significant variance between expected and actual losses.
Preparers should expect heavy scrutiny of these variances from auditors, regulators, shareholders, and other stakeholders who will have the benefit of hindsight in critiquing an entity’s estimates and underlying process. The processes put in place during the implementation of the CECL standard should include analyzing variances to address why those variances occurred and their potential implications to the forecasts used to estimate credit losses.
It is critical to understand the underlying drivers of actual activity relative to expectations to support an assertion that the model is designed properly and operating effectively. Not being able to address the reasons for the variances could lead an auditor to conclude that internal controls are not functioning properly, which could trigger a restatement and increased shareholder litigation. Performing sufficient analysis to understand the underlying drivers of differences creates confidence in management’s ability to identify the business risks, along with how those risks will change in response to economic forecasts, that should be captured in measuring expected future losses.