Bloomberg Tax
Free Newsletter Sign Up
Bloomberg Tax
Free Newsletter Sign Up

The Wild Ride of 2020 for CECL—Investors and Preparers Hope for a Typical 2021

Dec. 31, 2020, 9:00 AM

As 2020 comes to a close, so does the first year of application of the current expected credit losses (CECL) impairment model for most public companies. As with the implementation of most major accounting standards, companies anticipated and planned for the challenges in applying CECL. However, these best laid plans were complicated by the uncertainty caused by the Covid-19 pandemic, which upended much of the world in the first quarter of CECL’s application.

Since its issuance in 2016, companies most impacted by the credit losses standard spent significant time and resources developing controls and processes to capture and utilize historical data and analyze current and forecast future economic conditions in order to estimate future lifetime credit losses. Financial models were built to forecast credit losses and were calibrated based, in part, on observed relationships between loan data, macroeconomic variables (such as the U.S. unemployment rate or gross domestic product) and historical credit loss experience.

Controls and processes were put in place to establish a level of corporate governance expected and required from public companies and regulated entities. Many preparers established implementation work plans that started early so that models, controls, and processes would be in a position to operate under a “business as usual” environment beginning in Q1 2020. Unfortunately, like almost every aspect of our lives, the Covid-19 pandemic and its economic impact had a dramatic effect on preparers plans. While companies had already determined the impact of adoption and recorded their transition adjustments, they weren’t able to complete one full quarter using the CECL model before the pandemic would test whether management’s judgments and processes were reasonable.

2020 has been anything other than “business as usual” and everyone’s daily routine is nothing like it was when this year began. Employees had to pivot and adapt to a remote working environment, resulting in a need to ensure controls and processes could still work effectively. The data points and macroeconomic forecasts fed into the credit loss estimation models were stressed beyond the levels to which models were calibrated and forecasts were changed dramatically and frequently given the uncertainty and evolving views of the future. Governments around the world reacted by providing an unprecedented amount of economic stimulus, such as direct payments to individuals and/or enhanced unemployment benefits and programs to support businesses, which resulted in a disconnect between the models correlation of expected credit losses and macroeconomic data (e.g., U.S. unemployment rate). Companies had to react quickly and decisively in all aspects of their business, including when it came to estimating future credit losses. For a number of financial assets, expected future credit losses is a complex highly judgmental management estimate even in stable economic times.

While all companies need to estimate credit losses under the CECL model, similar to other accounting and reporting matters, they have different portfolios of assets, so may use different policies, process, controls, and quantitative models to generate the estimate. As a result, there is not a “one size fits all” solution in adapting estimates to the economic environment. Some preparers adjusted quantitative inputs into (or correlations within) their financial models. Many companies utilized qualitative overlays to adjust the output from credit loss models to address current conditions and economic forecasts not considered in the models. Regardless of the methods used to adjust their estimates, the shared goal remained: developing a reasonable estimate of expected credit losses.

With the benefit of hindsight, the allowance for credit losses estimated under the CECL model generally increased over the first six months of 2020, but at many institutions the allowance remained relatively “flat” in the calendar year third quarter. In certain asset classes, credit losses estimated by companies have not yet been “realized,” in part due to loan modification and deferral programs, government stimulus, and other factors. What has been reinforced over the first half of the year is the need to have appropriate documentation on how the allowance for credit losses is determined. Not only from a quantitative perspective, but equally as important, from a qualitative perspective.

As previously mentioned, significant adjustments, whether at the front end or back end of the models, took place over the reporting cycles, meaning that the delineations between what was a purely quantitative result and what was a qualitatively adjusted result became less clear. As a result, the documentation regarding these adjustments has become more important since they may have had a significant impact on the allowance for credit losses reflected in the financial statements. In looking ahead to the end of the calendar year, while there is much to be hopeful in the world’s response to the virus, much uncertainty remains, including the timing and speed of the forecasted economic recovery, which may differ based upon individual facts and circumstances.

By year end, calendar year public companies will have had three quarters of experience in developing the CECL estimate in an uncertain economic environment and with a remote work environment. They have taken the lessons learned over these reporting cycles and applied them to improve their related processes, controls, and models. Just because the original models developed to estimate expected credit losses under CECL may not have been calibrated for the dramatic change in the current economic environment, this does not mean that they are not fit for purpose for the “regular” environments in which they were initially designed to operate. Looking to 2021, many hope that economic forecasts and inputs will revert to the levels and scenarios contemplated when the models were designed, which may lessen, but not necessarily fully eliminate the need for qualitative adjustments to the model’s output.

Throughout 2020 there has been an evolution of disclosures relating to the new credit loss estimate. This is typical of newly-issued accounting standards as companies gain experience, have an opportunity to review the disclosures of their peers, and discuss financial information with users of their financial statements, including investors. CECL, coupled with the impact of the pandemic, is no exception. Disclosures on inputs into estimates, including macro-economic variables, and information on the use of multiple scenarios have evolved throughout the year. In addition, disclosures around credit risk management strategies, loan deferral programs, and the results of those programs have evolved and been well received by users of financial statements.

Disclosures, with a focus on helping users understand how the allowance for credit losses has changed, are expected to continue to evolve in 2020 annual reports and into 2021. The FASB has discussed disclosures and a few other topics based on the feedback received as part of the very early stages of their post implementation review of the new credit losses standard. Investors have articulated that information they received outside of the financial statements, such as quarterly earnings materials, played an important role in understanding what the key drivers for changes in the estimate were across reporting periods. Additionally, the lack of consistency and transparency in the CECL disclosures has made comparability challenging. This is not completely unexpected given that the CECL estimate is highly judgmental and complex in a benign environment, and even more so during a pandemic.

We expect increased focus by preparers and users on the CECL disclosures, and investors will likely seek ways to increase comparability across sectors. As companies prepare to close out year end results and the first chapter in the new credit losses guidance, they are well served by continuing to focus on providing decision useful information to users consistent with the intent of the CECL model.

This column doesn’t necessarily reflect the opinion of The Bureau of National Affairs Inc. or its owners.

Author Information

Thomas Barbieri is a partner and Deputy Chief Accountant in PwC’s National Office.