Shifting the mindset about financial model validations

At banking organizations, financial model validations can be simply viewed as a necessary task on a checklist for following regulatory guidance. Some institutions also believe that the quality of a model validation is less important when the institution or business line is successful and when local, regional and national economies are all thriving.

At moments like this, the model risk management consultants at MountainView Financial Solutions, a Situs company, remind institutions that periodic model validations are not just tasks for “identifying potential limitations,” as stated in one sentence of Statement SR 11-7 from the Federal Reserve. While periodic reviews are mandated, MountainView believes that financial model validations should be more than simply complying with regulations and managing risk; instead, they can be an opportunity to improve performance.

While all regions of the United States are currently experiencing economic prosperity, SR 11-7 points out that model validations can be more critical now than at other times. “Validation is an important check during periods of benign economic and financial conditions, when estimates of risk and potential loss can become overly optimistic and the data at hand may not fully reflect more stressed conditions,” according to the statement.

MountainView has seen more banks and credit unions embrace these two philosophies, as the number of model validations the company provides continues to increase annually. Performance improvement – and the strategic decision-making confidence that comes from it – has also been a driving factor in the increasing number of validation requests MountainView has received for non-traditional models such as mortgage servicing rights, mortgage pipeline, funds transfer pricing, and Allowance for Loan and Lease Losses, to name a few, according to Christine Mills, Managing Director and Head of Model Validations at MountainView.

MountainView’s independent model validations follow SR 11-7 guidance by including a conceptual soundness review, ongoing model monitoring, model performance testing, outcomes analysis, and a review of the model governance solution. Consistent with the evolving perception of validations improving performance and not just being mandated, MountainView’s clients are also increasingly seeking to align with preferred modeling practices.

Offering an example of this request, Mills says, “Our clients are interested in knowing how their assumptions and policy limits compare to their peers, and this is provided in our validations as a benchmarking of their results to their peers.”

Mills adds that this is only one example of how perceptions are evolving across the banking industry. “Whether it’s the advent of fintech or something else, we’ve been seeing less of a ‘check the box’ mindset about the necessity, quality and frequency of model validations.”