DFAST and CECL: Analyzing the Impact of Potentially Adverse Outcomes

The recent rollback of banking regulations included an easing of stress testing requirements, including eliminating required annual stress testing for banks with less than $50 billion in assets. Still, interagency guidance states that all institutions should plan for ways to meet funding needs under stressed conditions, and annual stress testing should be an essential part of your risk management practice. Not only will stress testing reveal how your institution will perform under potentially adverse conditions, it will provide much of the information you need to conduct your allowance estimations under CECL and can be leveraged in your transition to and implementation of a CECL methodology. 

The intent of and impetus for DFAST and CECL are essentially the same. From the broadest perspective, both are about getting a better understanding of your portfolio to help you make better, more informed business decisions. Both grew out of the financial crisis and the realization that there were blank spots in risk management that needed filling with forward-looking information. DFAST looks primarily at adverse scenarios to see how the institution can handle them, as CECL considers the impact on your loan portfolio of future conditions, allowing you to create a reserve sufficient to handle potentially adverse conditions. If there is an expectation of adverse conditions, you will be able to apply what you learn from DFAST to your CECL process. 

In our presentation at the MST 2018 National CECL Conference, we detailed the similarities and differences between the two regulations, including:

• DFAST is prescriptive. Scenarios are provided including a specified period of time over which to test performance. CECL is not prescriptive and the period under the microscope is the life of each loan.
• As such, CECL is restricted to the loan portfolio, whereas DFAST extends to all the institution’s financial instruments.
• In considering qualitative factors, DFAST puts more weight on macroeconomic and environmental factors, whereas CECL encourages the institution to look closer to home, how local market factors will impact the analysis.
• And of course, while DFAST is a supervisory activity, CECL is an accounting activity.

Both activities emphasize the importance of portfolio segmentation. A stress testing process wants you to consider parts of your portfolio as inherently vulnerable, like those already with a certain level of risk and volatility, and those that can create vulnerability, such as a particularly large loan that were it to go bad would adversely affect your entire portfolio. For either stress testing or CECL, you should be able to stratify those segments in your portfolio, and your CECL segments should be at least at the level of granularity as your DFAST segments. 

The point is that you don’t have to start from scratch for CECL if you’ve already done your work for stress testing.

Talk to MST about assisting your financial institution with CECL preparation.

About the Author

Chuck Nwokocha is a Senior Advisor for MST Advisory Services, assisting financial institutions nationwide in the interpretation and application of Current Expected Credit Loss (CECL), the new federal allowance accounting standard. His professional experience extends to various disciplines, including operations management, sales, human resources, product development and management in software licensing and delivery (SAAS).

Chuck’s extensive work as a consultant to financial institutions in risk management includes:

• loan portfolio management, focusing on the Allowance for Loan and Lease Losses (ALLL),
• management of stress testing, credit analysis, risk rating, and loan administration management.
• infrastructural change management and the implementation of the related technologies

Chuck is frequently called upon for commentary on various industries and issues, as well as to provide counsel through webinars and whitepapers, including such subjects as:

• credit risk modeling
• loan segmentation
• reserves calculation