Preparing for CECLImplementation

CECL implementation is approaching, and there are some new terms to become familiar with- historical lifetime loss experience and reasonable and supportable forecast. To aid in understating the new concepts, take into observation this high-level approach reflected in the equation above. The new allowance for credit losses (ACL) under CECL is determined by these three main factors: historical lifetime loss experience, adjustments for the current environment, and adjustments for reasonable and supportable forecasts. These terms and content consideration are outlined below. Additionally, there are some other implementation changes, which may need to be considered in existing practice, including but not limited to:

Purchased credit deteriorated loans (formerly purchased credit impaired loans)
Credit policy options (collateral dependent loans, fair value option)
Unfunded loan commitments

Historical Lifetime Loss Experience

Historical lifetime loss will represent an expected lifetime loss rate verses the current annualized average. This historical information should be captured in a format that allows analysis like an excel spreadsheet. Remember, CECL is scalable to the size and risk profile of the bank to meet the requirement of an analysis of the financial institution’s loss history and factors; a complex modeling instrument is not required. Here are some thoughts for the gathering of loss data:

Original balances (not renewal balances) must be gathered for starting the process of estimating the historical lifetime of a loan.
Consider how loss history data is captured and presented; most commonly represented as charge-offs and recoveries.
Consider how renewals and extensions appear in the financial institution's data (is the loan a renewal or new loans?).
Historical loss factors must be relevant to the current portfolio – don’t apply historical loss factors derived from commercial real estate to auto loans.
Organize the historical loss data by category (residential real estate) and life of loan (10 year contractual loans are best represented by 10 years of loss data). If the financial institution lacks that data; it should start capturing it now.
Consider prepayments and segmentation of pools by risk.

The segmentation of the financial institution's loan portfolio into pools by risk can be by similar risk ratings, call report categories or by other relevant credit loss information (FICO scores, loan-to-value and debt service ratios). CECL allows financial institutions to pool the loan portfolio based on the risk appetite. If a financial institution has a large concentrate of lending in a type of loan (i.e. land development), geographical location, or even to certain customer they can elect to create a separate pool to capture more granular loss data.

Adjustments for Current Environment

Qualitative factors (q-factors) are still relevant under CECL. As with the current accounting method, historical experience may not fully reflect an institution’s expectations about current events. Therefore, the financial institution should adjust historical loss information, as needed, to reflect current conditions. The analysis must be appropriate to the size and complexity of the institution and complex modeling techniques are not required.

Adjustments that are made by management must be reasonable, adequately documented and relevant to the portfolio. Management should consider relevant qualitative or environmental conditions that may cause the financial institution's current estimated loss experience to differ from past experiences. Some examples include:

The nature and volume of the institution’s financial assets.
The volume, trend and severity of past-due financial asset(s), nonaccrual asset(s), watch, special mention adversely classified or independently graded asset(s), work-outs and restructurings.
Changes to the underlying value or liquidity of collateral on financial assets.
Management’s lending policies and procedures, including changes in underwriting standards, collections, write offs and recovery practices.
The quality of the credit review system.
The experience, ability, and depth of lending, investment, and collections management, as well as other relevant staff.
The effect of other external factors, such as competition and the regulatory, legal and technological environments.
Actual and expected changes in the general market condition of either the geographical area or industry in which the institution has exposure.
Actual and expected changes in international, national, regional, and local economic and business conditions and developments that affect collectability of financial assets, including the condition of various market segments.

Management may also consider the effect of the following factors on the collectability of the institution’s financial assets as of the reporting date:

The existence, growth and effect of any concentrations of credit.
The effect of infrequently occurring events such as natural disasters, fraudulent activities, or material data breaches.
Changes in underwriting trends such as loan-to-value advance rates, debt service coverage ratios, borrower liquidity ratios and volume of identified policy exceptions.

Examples of factors that are more likely to affect the collectability of debt security portfolios are listed below. This list is not all inclusive and all of the factors listed below will not be relevant to all institutions. Management may consider the following factors for debt securities as of the reporting date:

The effect of recent changes in investment strategies and policies.
The existence and effect of loss allocation rules, the definition of default, the impact of performance and market value triggers, and credit and liquidity enhancements associated with debt security instruments.
The effect of structural subordination and collateral deterioration on tranche performance for debt security instruments.
The effect of legal covenants associated with a financial asset.

Reasonable and Supportable Forecasts

The CECL standard acknowledges that management may not be able to develop forecasts for the entire contractual life of a financial asset. In those instances, management should rely on forecasts of forward looking information that are both reasonable and supportable and are relevant to assessing collectability of cash flows. The standard does not prescribe a specific method for determining a reasonable and supportable forecast nor does it include bright lines for establishing a minimum or maximum length for a reasonable and supportable forecast period. Judgment is required in selecting an appropriate period for each institution.

Reasonable and supportable forecasts may vary by portfolio or by individual forecast input. Forecasts may be based on data from internal sources, external sources, or a combination of both. Management is not required to search for all possible information nor incur undue cost and effort to collect data for its forecasts. However, readily available and relevant information in assessing the collectability of cash flows should not be ignored.

The standard does not require the use of multiple economic scenarios when developing a reasonable and supportable forecast. However, management is not precluded from considering multiple economic scenarios when estimating expected losses. The use of multiple scenarios, if applicable, may be included in the reasonable and supportable forecast or as a qualitative adjustment when estimating the ACLs.

For those periods beyond which a reasonable and supportable forecast can be determined, management should revert to historical credit loss experience, or an appropriate proxy. Management may revert to historical loss information at the input level or based on the entire estimate of loss. No specific reversion technique is required by the standard.

Changes in the level of an institution’s ACLs may not be directionally consistent with changes in the level of qualitative adjustments due to the incorporation of reasonable and supportable forecasts in estimating expected losses. For example, if improving credit quality trends are evident throughout an institution’s portfolio in recent years, and management’s evaluation of reasonable and supportable forecasts indicates expected deterioration in credit quality during the forecast period for its financial assets, the ACL level as a percentage of the portfolio may increase.