CECL for Community Banks: The Practical Path Recap
Abrigo Executive Risk Consultant Tim McPeak and Risk Consultant Danny Sharman recently shared practical tips for community banks facing the CECL transition. The session included a good overview of CECL expectations and followed up on points discussed in an earlier regulatory webinar.
All banks, including small, non-complex community banks, will have to comply with the new accounting standard change from an “incurred loss” notion to the “current expected credit loss” notion (CECL). The future guidance is non-prescriptive, which provides banks flexibility in determining how to approach the transition. Although more room to thoughtfully perform the calculation for the allowance for loan and lease losses (ALLL) can be positive, community banks may struggle with determining a clear path forward. The transition will require a lot of planning and preparatory work. Community banks need to start considering how CECL is going to change their current data requirements and ALLL calculations.
In a June 17, 2016 Interagency Joint Statement, regulators noted that “agencies do not expect smaller and less complex institutions will need to implement complex modeling techniques.” McPeak and Sharman reiterated this point but noted that moving from the incurred loss model to lifetime expected losses requires different inputs into existing models and may encourage some institutions to evaluate more sophisticated methodologies. Although software will not be necessary to assist with the change, it offers many advantages, McPeak and Sharman said. Every financial institution should consider what would be appropriate for its own size and complexity.
Abrigo consultants also discussed the importance of data, and how it is one of the central components of the implementation of CECL. Data is critical to segmentation election, the ability to execute even “simple” loss rate methods and forecasting. For all of these areas, loan level details will be important. It will also be required for community banks to know what historical data they have on hand. After this analysis, a community bank should recognize where there are gaps. Comparing available data to what is expected under CECL may seem like an overwhelming process, however, McPeak noted how it is important to remember that there is no such thing as a perfect set of data.
326-20-30-2) stresses that “pools should be as granular as possible while maintaining statistical significance”. McPeak and Sharman said it is important to be aware of over-segmentation, for it is a more common issue than a lack of granularity. To achieve statistical significance, loan counts matter more than the dollars in each segment, they added. One common red flag for segmentation is when there is a single loan pool with 1/3rd of a community bank’s portfolio in it. Additionally, a loan pool that may be “too small” has less than 25 or 50 loans. Before changing segmentation procedures, it was recommended that a community bank quantify its impact and test different approaches.
Learn more about navigating the CECL transition.
McPeak and Sharman outlined three different loss rate approaches: snapshot/open pool, remaining life, and vintage analysis. These approaches were the same methods presented in the February regulatory webinar due to their scalability for smaller institutions.
Snapshot/open pool: This approach has many names, such as cohort analysis and static/closed pool, but essentially, it is a snapshot of the loan portfolio, and it tracks that loan portfolio’s performance in subsequent periods until its ultimate disposition. It is used commonly for finding the average historical loss rate over some time frame and is fairly straightforward compared to other methodologies. The new CECL standard will require the “snapshots” to be followed for the contractual life of a loan, adjusted for prepayment tendency. For community banks, it may be troublesome to calculate for loans with longer life expectations because of a lack of historical data.
Remaining life method:This utilizes average annual charge-off rates like today’s incurred loss model. However, institutions must make “pay down” adjustments for amortizing portfolios and apply annual rates. Weighted average amortization/payoff rates may be difficult to estimate at the pool level. This will involve considering prepayments and renewals; which adds complexity to the calculation. Inputs and critical assumptions will be scrutinized when using the remaining life method.
Vintage analysis: This is a well-documented calculation and is a good way to understand a key lifetime loss concept. However, it may not be applicable to all loan types; for example, its use for revolving lines does not make sense. Data is a big issue in a vintage analysis. Often, the length of loan history through accurate origination and renewal dates can make the calculation process more complex. More issues can arise when loan data is picking up recoveries from the recessionary period. McPeak and Sharman said that although it may bring some challenges, a vintage analysis is still a good way for calculating ALLL; it is best for pools that are homogeneous in risk and term, with high loan counts.
On-demand Webinar: CECL for Community Banks: The Practical Path
On-demand Webinar: CECL Methodology Webinar Series
Whitepaper: A Practical CECL Action Plan Whitepaper