Many institutions engage in third parties to run their interest rate risk model. These models are robust and can be very complex. Can we depend on model results from an external party to capture an accurate picture of the institution’s risk exposure?

The use of good data inputs and supported bank specific assumptions in the model are necessary for accurate results. Here, we discuss the various exercises banks should undergo to ensure they are using an effective and sound model.

Interest rate risk independent review

Interest rate risk is the risk to earnings or capital from movement of interest rates. We attempt to model this by analyzing the banks financials and projecting the impact to earnings and capital for given time horizons over various rate scenarios. The accuracy of the results is dependent on data inputs and reliable assumptions.

It is essential for bank management to have a full understanding of how their model works and be able to interpret the results. Banking regulators insist an independent review of the interest rate risk process should be performed annually. The independent review process should include the following components:

  • Assess bank’s interest rate risk measurement process, especially integrity of data inputs
  • Review reasonableness of model assumptions (prepayments, non-maturity deposits, etc.)
  • Evaluate processes involved with making assumptions (loan prepayment analysis, core deposit analysis, etc.)
  • Assess appropriate range of scenarios has been considered to capture risks under plausible and stressed financial conditions
  • Backtesting
  • Obtain third-party model validation

The independent review can be performed in-house or by an independent third party. As mentioned above, banks should obtain validations provided by vendors. Due to the high degree of customization for each institution, the bank should validate institution specific assumptions and data inputs in the model. A reconciliation of the backtest report should be performed as well. This exercise verifies projected and actual numbers were pulled from the interest rate risk reports.

The findings of the review should be reported to the board. The review provides the board with an evaluation as to whether the policies and controls over interest rate risk are being enforced. An effective review should give the board assurance the model provides reliable results and can be used for decision-making purposes. If changes are made to the interest rate risk model, this should be documented and updated in the bank’s asset-liability management (ALM) policy.

Backtesting measures and results

Backtesting compares actual outcomes with model forecasts. A backtest should be obtained or prepared at least annually, and we typically see this completed with end-of-year data. This is conducted over a 12-month period because a shorter period will not capture errors in the model. The projected rate scenario most closely resembling the economic conditions during the 12-month period should be applied. For example, if market rates were unchanged during the period, the base case scenario would be used.

There are three common backtesting measures; rate variance, volume variance, and mix variance. Given that most interest rate risk models use a static balance sheet, the primary focus is rate variance. The rate variance relates to the impact of rising or falling interest rate levels on net interest income. This measures the impact of the assumed repricing rate from one year ago for each balance sheet category and its impact on net interest income. Positive results suggest the repricing was not set high enough one year ago to reflect what actually occurred. Negative results suggest rate repricing assumptions were too high in comparison to what actually occurred.

The backtest report also includes volume variance and mix variance. Volume variance relates to the growth or decay of an account balance and its impact on net interest income. This measures the year-to-year changes of the balance for each category based in proportion to the current total earning assets balance and its effect on net interest income. Mix variance refers to the actual composition of the balance sheet. Mix variance differs from both rate and volume variances, mix variance is concerned with the relative proportions of balance sheet items to one another. For example, review whether a large rise in real estate loans generated a shift in the percentage represented by real estate loans or if all the loan accounts grow by a similar degree.

The results of these measures help identify any areas in the model that need to be addressed. If outcomes fall outside a range or statistical confidence level it would require further investigation and documentation. For a majority of banks, the 2023 backtest results show a sizeable rate variance due to deposit accounts. Banks continued to raise rates on deposit accounts to retain and/or attract customers even though the increase in market rates slowed down.

By identifying outliers, we can adjust the model as needed to improve the accuracy and reasonableness of the bank’s interest rate risk model.

Model validation

The bank should obtain documentation showing a credible independent third party has performed a validation of all model mathematical calculations and software mechanics. The purpose of the validation is to test the fundamental theory, calculations, and operational capabilities of the model to verify the model is performing as expected. An effective validation ensures a model is sound, and outcomes can be used for management to develop and execute business strategies. The exercise will also identify any deficiencies or potential limitations of the model and the impact on the results.

The review should be conducted by individuals independent of the model development, management, or monitoring process, in accordance with regulatory guidance. Due to the complexity of the model, the individuals who perform the validation should have the knowledge and technical expertise to complete the review.

The validation should include an executive summary, scope, test and results, findings, and recommendations. The scope of the review might include the following components:

  • Interest rate curve construction
  • Stochastic modeling
  • Scenario application
  • Loan modeling
  • Credit modeling
  • Prepayment modeling
  • Deposit modeling
  • Cash flows
  • Derivatives

The review will run model tests and compare outcomes with industry leading vendor model outputs, which is also known as benchmarking. If available, some components can be reviewed internally by replicating the process with spreadsheets. This will establish the model performs as intended and the model estimates are in line with the expectations. When material changes are made to the model or a new version is adopted, a validation should be performed. The full validation process should be performed at a fixed interval and presented and reviewed by management. Discussions related to findings, recommendations, and action plans should be well documented.

In summary

It is essential for banks to monitor these complex models to ensure the outputs are sound because management depends on these outcomes to make critical decision for their institution.

To effectively manage model risk the interest rate risk independent review, backtesting, and validation process should be completed to show the validity of the outputs and uncover inefficiencies to be addressed.

The UMB Financial Services Group can perform an independent review, backtest, and validations for your institution. Please reach out to our team of analysts with questions.

Learn how UMB Bank Capital Markets Division’s fixed income sales and trading solutions can support your bank or organization, or contact us to be connected with a team member.

Disclosure

This communication is provided for informational purposes only. UMB Bank, n.a. and UMB Financial Corporation are not liable for any errors, omissions, or misstatements. This is not an offer or solicitation for the purchase or sale of any financial instrument, nor a solicitation to participate in any trading strategy, nor an official confirmation of any transaction. The information is believed to be reliable, but we do not warrant its completeness or accuracy. Past performance is no indication of future results. The numbers cited are for illustrative purposes only. UMB Financial Corporation, its affiliates, and its employees are not in the business of providing tax or legal advice. Any materials or tax‐related statements are not intended or written to be used, and cannot be used or relied upon, by any such taxpayer for the purpose of avoiding tax penalties. Any such taxpayer should seek advice based on the taxpayer’s particular circumstances from an independent tax advisor. The opinions expressed herein are those of the author and do not necessarily represent the opinions of UMB Bank or UMB Financial Corporation.

Products, Services and Securities offered through UMB Bank, n.a. Capital Markets Division are:

NOT FDIC INSURED | MAY LOSE VALUE | NOT BANK GUARANTEED