What Institutions Ask First

The right frequency depends on your institution’s size, balance sheet complexity, and the model itself. Some models are more challenging to run correctly and may warrant more frequent review. For most institutions, a validation should be performed no less than every three years. Larger institutions or those with elevated exam stress should plan for an annual validation. When significant changes occur — including a new model, a major assumption revision, or a material shift in balance sheet composition — a validation should be considered regardless.

If an examiner told you one was needed, the answer is yes. For others, the case is still strong. A validation often surfaces areas where the model could be used more effectively, including additional stress tests or what-if scenarios. It might also find that policy limits are too wide, or that back testing or a core deposit study would strengthen your process. In our experience, there are almost always areas for improvement, regardless of how long the current process has been in place.

In practice, this tends to be prohibitively expensive and rarely produces meaningful value. Running data through a second model almost always produces similar risk profiles. What the process generates is not useful insight. It generates a reconciliation exercise. The work centers on identifying why results differ and adjusting assumptions until the two models converge. That is not a good use of your resources.

There are no shortcuts here. A validation covers everything your institution does to measure, monitor, and report on interest rate risk. We review balance sheet segmentation, account categorization, and treatment of complex investments. We examine every assumption, including replacement rates, repricing sensitivity, prepayment speeds, discount rates, and non-maturity deposit retention. We review your policy and limits, assess model adequacy, and review ALCO and board minutes. We also comment on ancillary reporting such as back testing, assumption sensitivity analysis, and what-if scenario runs.

It depends on the situation. If a full validation was recently completed and assumptions have since changed, a focused review of those changes can make sense. If you are looking for a quick review every few years to reduce cost, we would likely not be providing worthwhile service. If you are unsure, give us a call. A brief conversation will clarify the right path forward.

You will receive a written report specific to your institution and our findings. It is not a template. After delivery, we will meet to walk through findings and answer questions. If you believe we have misinterpreted something, we want to know and will update the report accordingly. The goal is a report that is accurate, fair, and useful.

Yes. Examiner comments on ALM can cover model adequacy, assumption support, policy limits, governance, and documentation. Interpreting what examiners are asking for, and responding effectively, requires experience on both sides of that conversation. There are very few examiner concerns in this space that have not come up before. Give us a call and we will talk through what you are dealing with.

Yes. I do not sell an ALM or CECL model, I do not perform financial audits, and I am not affiliated with any model provider, software vendor, or accounting firm. My only service is independent model validations.

Most validation providers are either consulting firms that also sell their own ALM model or accounting firms where validation sits alongside audit and tax work. I am independent of both groups, which means no competing software to promote and no audit relationship to protect. The validation report you receive reflects what the model actually does and how it compares to regulatory guidance and examiner expectations.

Privacy Policy