“Oh Bother,” Another Model Validation

“Oh Bother,” Another Model Validation

Do you ever have quotes pop into your head that seem to just fit your situation? What about when the exam list or a model validation request makes its way to your inbox? I know what I hear it in my head, “Oh Bother!” I never fully appreciated this phrase from Winnie the Pooh until I was an adult, but his words ring true. “Oh Bother” encompasses the feeling of dread and disappointment. It dredges up the complete lack of interest in completing another model validation…and possibly some cravings for comfort food. For the overwhelmed bear, Christopher Robbin was always there to bolster his courage or to give another perspective, or maybe just some honey. So let me be that voice of encouragement for you today if “Oh Bother” is where you are with a model validation.

Yes, a validation is part of your regulatory exercise, but it also ensures your model and associated processes are operating at their full performance potential.  With earnings under pressure, lower capital ratios, rising inflation and excess liquidity (“Oh Bother” appears to apply here as well!), it is more important than ever to receive the best information possible to assist in making key strategic ALM decisions. Your next validation can help set the credit union up for success, so maybe it’s not as much of a bother as it seems.

To make this more appetizing (like a big pot of honey), let’s break it down into simple steps.

Selecting a Vendor. Choose someone who has the expertise and experience in interest rate risk and interest rate risk models and is also independent of the credit union’s modeling process and individuals responsible for modeling. Ask for bios of staff that will be assigned to the validation and the person performing the final review of the report. Are they familiar with your model?

Preparing. Take time to review the source data files, the model’s chart of accounts, model assumptions and reporting provided to the board and ALCO. Review your policies and risk limits and make sure they align with your reporting and risk measurement. If you outsource your modeling, set up a review session to freshen up your understanding. Start with the right intent to make sure things are in good order before the process begins.

It also helps if you know what to expect with the testing involved in a validation. The key testing components of an interest rate risk model are as follows:

  1. Model set up ensures that the credit union’s various accounts are properly defined in the model’s chart of accounts to effectively model specific behavioral characteristics. In performing model validations, it is not unusual to find fixed and adjustable-rate loan products or non-maturity share accounts with different dividend rates or pricing tiers aggregated together into a single account.
  2. Integrity of the current data loaded into the model is critical to the output results. Testing in this area ensures source data files contain required product attribute fields, the model accurately reflects contractual terms and balances and repricing data of source data files and model balances reconcile to the general ledger. One of the most common exceptions found in validations is missing data for adjustable-rate loans.
  3. Assumptions within an interest rate risk model include new volume reinvestment attributes, prepayment and other option-related risk, non-maturity share behavior and net economic methodology. A validation reviews the basis used to determine each of the critical model assumptions, as well as the reasonableness of the assumptions, and ensures that the assumptions are properly employed within the model.
  4. Output of results is the final model testing component. This includes a review of the resulting model reports to ensure they reflect expected outcome and the credit union’s policy limits. Since risk is measured under a range of rate scenarios, make sure that reports are easy to understand and support the decision making process.
  5. Policy review is often completed during the validation in addition to the model review. Third-party providers may also assess other areas of the credit union’s interest rate risk management process such as policies, oversight and internal controls.

Utilizing the recommendations should be the final stage of the validation process with your ALCO. The credit union should review each of the recommendations with the vendor, decide upon the disposal or implementation of each of the recommendations and submit the validation report along with management’s recommendations to ALCO.  Remember, the goal is to ensure that you have information that you can rely on to make informed balance sheet decisions during this uncertain rate environment. Keep in mind that not all recommendations are a good fit for your model, so be sure to implement with intent and discard the other items that are not as impactful to your decision making.

If you are feeling unsure of the process, don’t be bothered. Instead, take it as an opportunity to improve your ability to forecast and rely on your results by leveraging an existing tool to its fullest. During this quickly-changing market environment, we need to take advantage of every tool possible to aid our success. The next time you hear “Oh Bother,” try to utilize this process of understanding to break down the steps and lead you to an easy validation, and maybe a smackerel of honey to help you along the way.

Lisa Boylen is a senior ALM consultant for Vizo Financial. Lisa is responsible for managing and providing ALM reporting, modeling, validation and consulting services for credit unions. She is also responsible for providing on-going support and education to ALM users, management and their boards.

Melissa Scott serves as Vizo Financial’s vice president of ALM services. In this role, she is responsible for managing and providing ALM reporting, modeling, validation and consulting services to credit unions. She is also responsible for providing ongoing training, support and education for ALM users, management and their board of directors. She also holds the designation of certified public accountant (CPA) and is a member of the North Carolina Association of Certified Public Accountants.