The Different Use Cases for CECL Methodologies

Financial institutions are currently planning and building models for the quickly approaching implementation of the current expected credit loss standard, or CECL. The accounting change brings many concerns surrounding implementation dates, modeling, qualitative factors, economic forecasting, and documentation/reporting.

More specifically, many bank and credit union managers and executives think that finding the right methodologies to use when calculating their allowance is currently their largest concern.

Abrigo hosted its 7th annual Lending & Risk Summit in September 2018. CECL was a large focus at the conference, which included numerous presentations on the impact of the standard and discussions among bankers on what they have been doing to prepare.

Neekis Hammond, Managing Director of Abrigo Advisory Services, gave a presentation on CECL decision points at Summit and explained how different choices when building out a model can lead to either successful or unsuccessful results. In the presentation, he polled an audience of more than 160 presentation attendees to see what value they saw in using CECL methodologies for other functional areas at a financial institution.

Options for answering included that CECL methodologies were most valuable for profitability analysis, stress testing, valuation/exit price or internal meeting packages. Additionally, attendees could express that they felt all options were equally valuable or not valuable.

More than 56 percent of the audience felt that the options were equally valuable. The second most popular answer was stress testing, with more than 24 percent choosing stress testing to be the most beneficial functional area for use of CECL methodologies. Although banks and credit unions typically find choosing the right methodologies to be challenging, all presentation respondents thought CECL methodologies could be of value for different business initiatives.

“Vintage, PD & LGD, Migration, and DCF, to name a few, provide for valuable insight no matter the exercise. For CECL, it is prudent to make objective elections by observing various outcomes. Independent of CECL, deploying these models for risk, pricing and profitability, fair value, and/or stress testing can only improve portfolio management and inter-departmental symmetry. At the end of the day, side-by-side comparisons of various methodologies under multiple economic scenarios will prove to be one of the most valuable features of your CECL platform,” said Hammond.

Considerations for choosing CECL methodologies
Although there may be some additional positive use cases for CECL methodologies, institutions should still be careful in choosing the best one(s) for their loan portfolio. The standard is non-prescriptive on which methodologies should be leveraged. In order to make the best decision, Hammond recommended during his presentation, it’s best to start by analyzing one of the main pillars of CECL: data.

“Important CECL implementation considerations that can have big impacts based on operational realities include data and methodology selection,” Hammond said. “It’s vital for financial institutions to understand what can and cannot be done given the limitations of their data history and accuracy because certain methodologies are more reliant upon long-range historical data availability.”

Different methodology options available to financial institutions include migration analysis, probability of default/loss given default, vintage analysis, discounted cash flow, transition matrix and static pool analysis. There are numerous resources available to learn about these different methodologies, including what inputs are needed for each one and how to make the best decision based on what data is available to the institution.

“With methodology selection, let objectivity be the guide. Different methodologies yield different results; methodology decisions prior to observation may yield uncomfortable results,” said Hammond.