<img height="1" width="1" style="display:none" src="https://www.facebook.com/tr?id=815305791870634&amp;ev=PageView&amp;noscript=1">

Echo Partners Community Bank Blog

    FASB Clarifies CECL

    Flexibility direction sign with sky background

    FASB’s 2nd Q&A reiterates that CECL is designed to be both scalable and flexible. Each individual Q&A topic emphasizes this approach.

    Forecasts for expected future conditions do not require statistical or computer modeling. FASB uses Q-factors as an example of how to accomplish this without modeling.

    Historical loss information alone is not sufficient for CECL. It must be modified and amplified by adjustments to reflect current and expected future conditions.

    Historical loss period selection should be used to best estimate credit losses. It need not include all loss periods, nor even require sequential periods. Flexibility is the hallmark of the standard.

    Financial institutions are not required to consider all sources of information when estimating CECL. Relevant information that is reasonably available without undue cost and effort is appropriate.

    If internal sources of information are more relevant, FASB clarifies that external sources are not required. The entity may decide to use external or internal information based on the specific circumstances.

    Bottom line is that virtually all aspects of CECL may be revised as needed to best estimate credit losses.

    How to Calculate CECL

    Portrait of a handsome male chef cook reading recipe book while preparing food isolated on a white background

    It’s easy to calculate your own aggregate CECL report.

    1. Obtain data. Bank or credit union, start with the regulatory Call Report/5300. Get all data since 2007 for all institutions.
    2. Organize it using the default loan segmentation. Convert YTD data to quarterly and fix any errors.
    3. Calculate loss rates using your chosen methodology for every segment and possible loss period. Aggregate the losses relative to the beginning loan balance. Remember you can’t lose money until after you’ve made the loan.
    4. Calculate peer data to obtain the same comprehensive set of loss rates for all segments. Determine how you are going to select, filter and use peers.
    5. Identify and organize a process for calculating, valuing and reporting CECL adjustments related to current and expected future conditions.
    6. Organize and establish a process for collecting and implementing Q factor adjustments. Be sure to make it consistent and symmetrical for both direction and magnitude.
    7. Run thousands of test cases each quarter. Track down any outliers and correct any errors.
    8. Obtain an independent 3rd party model validation opinion.
    9. Fully document everything and have an analyst ready to help with your questions.

    Yes it’s easy to calculate CECL but it’s even easier to get me to help you.

    Let me know if you would like a sample for your financial institution.

    CECL Loss Rate Periods

    close up of a calendar-1

    At different times over the credit cycle it’s likely that your bank or credit union will experience different loss rates. That’s important when you select CECL credit loss periods.

    Consider 2 examples.

    You could use all of your loss rate history (your "all periods" experience) from 2007 to present. There’s a lot to recommend this approach. It contains the most periods of credit history. It arguably includes a full credit cycle. It also includes the aftermath of the Great Recession/credit crisis.

    But a lot changes in a decade. That older loss experience might not accurately represent your credit experience today.

    Another approach might be to focus on your most recent loss rate periods. After all the most recent periods likely contain loss data that’s more highly predictive than older data. It’s certainly possible that your next period of credit losses might look a lot more like the current period than ancient credit history.

    The problem here is that we’ve been in a period of relatively benign credit experience based on a continuing expansion. Things won’t always be like this and we know it.

    Consider using both your all periods and your most recent loss rate periods. Blending these 2 together can give you the best of both worlds…Robust credit history tempered by current experience.

    CECL Q Factors

    Close up of hand repairing pipes with wrench-1

    With all the CECL attention focused squarely on methodologies and loss rates, Q factors seem to have retreated back into the shadows. But Q-factors are an in important part of a CECL process.

    Q factors are a regulatory staple. We’ve used them since the Interagency Policy Statement on ALLL way back in 2006. And our regulators are in unison again that Q factors will remain relevant with CECL.

    The question is how you will best use them.

    Most of all, Q factors are a natural part of adjustments related to changes in current conditions for purposes of CECL. I suggest you build upon them from there.

    Designed to bring qualitative information and management judgment into the ALLL process, be careful you don’t go overboard using them in a CECL context. Specifically don’t use them in a way that begs your examiner to charge you with a “too subjective” process.

    I recommend using Q factors with a predetermined adjustment value matrix that is consistent with respect to both direction and magnitude. In other words make your qualitative adjustments as quantitative as possible.

    And be sure to limit the value of Q factor adjustments. Adjustments are the seasoning to the loss rate steak. Don’t let them overpower your loss rates.

    Expect CECL Reserves to Increase

    Close up of woman mechanic with yellow helmet in hand against city background

    The simplest possible result from CECL is one that I’ve seen many community bankers resist. And that is the idea that reserves under CECL should grow.

    Today community bankers primarily use annual loss rates to calculate reserves. CECL requires us to estimate lifetime losses and reserve accordingly. Here’s a simple example borrowed from our regulators…

    Assume you make 3 year loans. Your annual charge off today is based on 1 year, while your CECL lifetime charge off would be based on 3 years. It’s just common sense that the three year loss rate would be higher than the one year loss rate.

    But I can hear you say, you told me that my overall total loss under CECL will be the same as under today’s incurred loss method. How do you reconcile a higher reserve with the same overall loss?

    Reserve building is a timing difference. CECL doesn’t increase our loss. CECL forces us to reserve our loss earlier than we otherwise would have.

    Most of us will have a significant “Day One” reserve building requirement when we adopt CECL. And unless you are a credit union, your regulator has already provided relief in terms of a 3-year phase in of negative Day One CECL impacts on capital.

    Don't be afraid of CECL. But if you have CECL questions let me know.

    CECL and Peer Data

    Young businessman having a serious discussion with two female co-workers as they stand together discussing a hand written flip chart

    There’s a lot of talk about CECL and peer data. When you get right down to it there are 2 ways that peer data can assist in the CECL process.

    1) Filling in missing loss rate data.

    With many community financial institutions having pristine credit experience there’s the likelihood that some credit loss rates may report as 0% (zero).

    While zero losses are always a great accomplishment, you are likely not going to get a lot of support from your examiner in expecting zero losses to go on forever, or even for the expected lifetime of new loans. That’s where peer data comes in handy.

    2) Helping with adjustment support.

    You have adjustments for both current and expected future conditions. It’s not hard to imagine using peer data to help support a slight increase or decrease from your historic loss rates.

    Both of these methods depend on reference to many peers. The problem with peer data for CECL is that this is not the way bankers are used to using peer data.

    CECL peer data should be based on large numbers of institutions to get a representative sample, not a targeted handful of local banks or chosen competitors. You’re looking for different experiences, not more of the same.

    Are you using peer data in your CECL process?

    CECL Vintage Method

    Colorful bawling ball waiting to be chosen

    One of the most common loan level approaches to CECL is the vintage method. But at its core vintage is remarkably like the aggregate loss rate methods.

    Vintage performs the same basic loss rate calculation on individual loans and then segments them according to origination date as well as loan type.

    These individual loan loss rate experiences are collected in a table where the timing of each loss or recovery is associated with the time from origination at which it occurred. Then after aggregating these experiences an average loss rate pattern is developed for each loan type based on the origination date.

    From this table estimates of likely future loan loss are developed for each loan segment, allowing us to prepare a CECL reserve.

    The surprising thing about vintage is that this process might not improve or reduce your CECL estimate.

    Value from vintage relies on 2 components (one of which isn’t vintage at all).

    1. Improved Segmentation. Improved segmentation may improve the quality of your CECL estimate.
    2. Discounting. By identifying the vintage cash flow you can use the effective loan rate to present value the expected loss. But that’s really DCF bolted on the back of vintage.

    Let me know your CECL questions.

    Community Financial Institutions and CECL

    Free Estimates card isolated on white background

    Community Financial Institutions and CECL

    Recently I’ve spoken with several community financial institutions about CECL. Here’s what I’ve heard:

    • We don’t have the time to invest in CECL. This includes no time to learn new software. Community bankers always wear many hats, but the downsizing experienced recently has pushed them to their limits.
    • We don’t have the resources to invest in CECL. Community financial institutions have been getting squeezed by new and enhanced compliance measures for a decade. They’re running out of bandwidth and need a cost effective solution.
    • We don’t have the expertise to implement CECL. While many CECL concepts are similar to current ALL practices, the assumptions and adjustments differ. There’s no one in house to pull it all together and educate the team.

    But opting out is not an option.

    CECL is happening, and even the smallest private institutions need to get going.

    But how?

    Start with learning more. Then estimate your CECL exposure (potentially a multiple of current reserves).

    I’d be glad to give any community bank or credit union an estimate of your CECL impact. Just send me a message or email to get started. No cost, no obligation. Just helpful info you can use.

    CECL: Aggregate or Loan Level?

    Office folder with inscription Methodology on Office Desktop with Office Supplies. Business Concept on Blurred Background. Toned Image.

    The biggest choice community banks and credit unions have with respect to CECL implementation is whether to use an aggregate or loan level approach. In many ways this boils down to size and data availability.

    Here are some thoughts to guide your choice:

    Size and complexity: Regulators have said CECL is designed to be scalable and that if you are using a loss rate method now you will likely not need a more complex method with CECL. Practically speaking most banks and credit unions less than $500mm assets should definitely consider an aggregate model. Larger or public institutions should look more towards a loan level solution.

    Data: Most aggregate methods rely upon data and segmentation as defined in your regulatory reports. That means the overwhelming majority of data gathering, segmenting and organizing is already done. Minimum data for a loan level vintage will include origination date/balance, date/amount of each loss/recovery, loan type code and mapping. I’d recommend you include effective yield to allow DCF. All data points must include specific loan ID to allow us to tie it all together.

    There is a final consideration: cost. Aggregate methods should be much less expensive than loan level.

    What makes sense for your bank or credit union?

    CECL vs. Incurred Loss

    frustrated with problems young business man working on laptop computer at home

    The difference between CECL and the existing incurred loss method is right there in their names.

    Our existing incurred loss method keeps us from booking losses into our reserve until a loss is actually incurred. In practice, the standard allows us to book losses when a negative credit event occurs that makes a specific loss probable.

    Losses are booked as incurred, and many financial institutions have learned to view incurred losses based upon an annual loss rate.

    CECL (Current Expected Credit Loss) on the other hand requires us to estimate the expected lifetime loss and book that loss at the origination of the loan.

    It just makes sense that the CECL reserve (estimating lifetime losses) will likely exceed the incurred loss reserve (aggregating incurred losses). But at the same time we must remember that the total loss recognized under either method will ultimately be the same.

    It’s hard to reconcile and understand the differences unless we consider the entire loan portfolio. The incurred loss method measures the current loss in the existing portfolio now, while CECL measures the lifetime loss, or the current risk in the portfolio.

    The key is remembering that the CECL loss is effectively equal to a multiyear incurred loss.

    Do you want to grow your bank profits with little to no risk? Click Here to  Discover How

    Written by

    Subscribe to Email Updates

    Recent Posts