As interest rates creep upward, federal regulators are increasing their scrutiny of banks’ asset-liability management (ALM) programs. ALM models typically focus on interest rate risk — though liquidity risk is also a significant factor.
To ensure that your bank meets regulatory expectations, here are five best practices to consider.
1. Get the Board Involved
In today’s high-risk environment, a bank’s board or asset and liability committee (ALCO) needs to take a proactive approach to ALM. This approach should involve understanding the risks associated with your bank’s products and activities, setting the proper tone at the top, and communicating the bank’s risk tolerance throughout the organization — often by adopting an ALM or interest rate risk (IRR) policy statement.
It’s important that your bank hire management personnel with expertise in managing risk effectively — and that it receives risk reports at a frequency appropriate to the bank’s level of risk, but at least quarterly. These reports should contain sufficient detail to allow the board or committee to understand the underlying assumptions, identify any noncompliance with bank policies and expose weaknesses in your bank’s ALM model.
2. Educate the Board
Directors need not be experts on ALM. But examiners expect board members to have a sufficient understanding of it to meet their oversight responsibilities and fulfill their fiduciary duties.
Banks need to provide board members with educational opportunities or include one or more outside directors on the board who have previous experience with balance-sheet risk management. Any outside consultant should attend board meetings to discuss report results.
3. Select the Right Model
Regulators expect banks’ ALM models to be adequate in light of their complexity and risk profiles. Formerly, for simple balance sheets, a maturity gap analysis may have been enough. This type of analysis measured repricing risk based on the potential “gap” in value between assets and liabilities that mature or reprice during a time period.
But more sophisticated times call for more sophisticated tools for banks whose balance sheets include underlying aspects of financial products or instruments. Typically, all balance sheets contain embedded options — such as prepayment options on both single maturity and amortizing securities or loans, withdrawal options commonly referred to as decay rates for deposits, put or call rights, interest rate caps, floors or fixed to variable or variable to fixed rate conversion rights — which demand more sophisticated simulation modeling. Evaluating the risks associated with these options requires detailed assumptions about future interest rates, repricing assumptions, economic conditions, and customer or investor behavior.
4. Evaluate Vendors Carefully
A number of vendors offer IRR and other ALM models that vary significantly in terms of complexity, data management and cost. When evaluating third-party models, thoroughly assess their ability to capture your institution’s risks. An interagency set of frequently asked questions (FAQs) on IRR management urges banks to consider the following:
- A product’s ability to model the bank’s current and planned on- and off-balance-sheet products. The model should support a level of data aggregation and stratification necessary to properly measure any highly structured instruments or institution-specific products.
- The model’s use of automated vs. manual procedures. The bank should consider whether the model has automated interfaces with bank systems, as well as the cost, hardware and software requirements, and necessary staff resources and expertise.
- The model’s level of transparency and the adequacy and comprehensiveness of vendor model validations (independent testing to ensure that the model is performing as expected) and internal control reviews (processes in place that allow comparison of the model balance sheet with the bank’s own financial reports).
Regulators also expect banks to have a reliable level of in-house knowledge of the types of IRR inherent in their balance sheets and the model’s capabilities to measure such risks in the event a vendor terminates a contract or goes out of business.
5. Document, Test, and Validate
To meet regulatory expectations, document all key assumptions incorporated into your ALM model, such as loan prepayment rates, core deposit decay rates and beta estimates (which measure how responsive management’s deposit repricing is to the change in market rates). Your bank also needs to conduct periodic back-testing to compare your ALM model’s projections to actual performance. This allows management to determine whether any assumptions need to be adjusted.
Don’t confuse back-testing with validation. Regulators expect banks to make sure their ALM models are appropriate for their risk profiles by obtaining annual independent validations. The validation process involves a review, by an outside expert, of the model’s logical and conceptual soundness. The expert also tests to determine whether the model’s mechanics and mathematics are functioning properly.
It’s likely that regulators will continue to scrutinize banks’ ALM programs and models. To avoid examination issues, banks should be proactive in making their programs adequately reflect their risk profiles. Contact your Elliott Davis advisor for more information and guidance regarding your ALM policies.
Sidebar: Revisit ALM when adding products or services
Banks should review and update their ALM policies before adding new products or services — or adopting new strategies. Additional modeling may be required to ensure IRR risk implications are identified. Too often, banks make these changes without sufficiently evaluating the risks, only to discover that management of those risks requires additional processes or resources that reduce their return on investment.
According to the FAQs, the failure of an existing ALM model to capture risks associated with new activities will likely be viewed as a management weakness, requiring corrective action.