Prospective CECL Methodologies: Transition Matrix

Part of a MST Blog Series examining prospective CECL-compliant methodologies

The road to CECL compliance ends in identifying the CECL methodology (or methodologies) that best suits your institution and your loan portfolios as well as complies with CECL guidance and your auditor and regulator demands. This MST Blog series is designed to help you examine prospective CECL-compliant methodologies. 

FASB has been steadfast in not specifying or even identifying all appropriate CECL methodologies. Because each institution is different, with different types of loan portfolios and different underwriting principles, the method or methodologies you choose must be determined by your unique blend of products, portfolios and markets as well as the quality and quantity of your available loan level data. 

That doesn’t mean you have to rebuild a methodology from scratch, only that you  must choose a methodology for each loan pool based on your judgment and circumstance. Most institutions have relied on historical loss analysis to estimate their allowance under the incurred loss standard, but under a rule that requires estimating loss based on assumptions that start on day one of the loan and follow it to its likely conclusion, an estimate based simply on what happened last year won’t fly – or comply. 

For insights on a Transition Matrix methodology, we turned to Chris Emery, senior advisor, engineering and MST director of special projects, who as a member of the MST Advisory team of consultants has been guiding lenders through the discovery and decisions that will lead to choosing and implementing a CECL methodology. 

What is a transition matrix approach to CECL?

Where an institution has applied a consistent risk rating system to its loans for at least a full economic cycle, it might want to use a transition matrix to track loan performance to estimate future losses, according to Emery. A transition matrix can provide a measure of probability of default by tracking, over quarter-end or year-end periods, how loans move, or transition, from one risk metric to another. The matrix can be used to follow credit scores or delinquencies, but risk ratings are most commonly used. 

How does the transition matrix process work?

•  Pick a rating period.
• Look at all the loans in that pool that existed during that period to see how they moved from one risk rating to another.
• Apply percentages to loans that kept the same risk rating, moved to another rating bucket, defaulted or were paid off.
• Determine the rate of default for each rating grade in the pool, which established the matrix probability of default for that grade.
• Apply the matrix to your current portfolio to project how those loans will move in the upcoming year and to termination of those loans – either paid off or defaulted. 

What are the key considerations for using a transition matrix?

• Consistency – A transition matrix is appropriate for any type of loan as long as the lender has consistently applied its risk metrics over several years. If ratings haven’t been applied consistently, or definitions for ratings or other rating criteria have changed over time, the matrix won’t produce reliable results.
• Sufficient rating data – Matrix calculations should cover a complete economic cycle, worse case scenarios as well as best, so that it can be a reliable forecast as economic conditions change.
• Establishing a transition matrix takes several years as the lender must follow the loans in the determined period and pool to their termination, default or pay off.
• Loss Given Default denominator – The matrix provides the probability of default part of the pd/lgd calculation. The loss given default must be calculated separately – there are several ways to determine the loss given default – that is, add the losses for the loans that have defaulted. The loss given default becomes the denominator; probability of default the numerator. 

For institutions with the required rating data history and consistency, a transitional matrix can provide a lifetime default rate for a particular grade for a particular pool, and a reliable and compliant method for forecasting future losses.

Considering Adjusted Annual Loss Rate, read the blog.

Read the blog on using Vintage as a prospective CECL methodology.

Read the blog on considering PD/LGD as a prospective CECL methodology .

For more on PD/LGD, read “An ALLL Methodology for Your Future”.

About the Author

Chris Emery has helped hundreds of financial institutions of varying asset sizes and employing all major core systems implement allowance technology that supports their efforts to comply with regulatory and accounting standards, including in their current transition to estimating the allowance under CECL. 

In addition to his client engagements with MST Advisory, Chris advises the technology group at MST which is charged with developing, implementing and supporting software solutions that assist financial institutions with calculating the Allowance for Loan and Lease Losses (ALLL) and now making preparations for CECL.