I’ve always said that deposit pricing (a/k/a “betas”), decay rates, and asset prepayments are “the big three” when it comes to interest rate risk (IRR) model assumptions. Why is that? Well, it’s because those are the ones that most people think have the greatest impact on model results, and, as a result, those are the ones that examiners tend to look at the closest. And while I don’t always agree that prepayments matter all that much (depending on the duration of any given institution’s intermediate- and longer-term assets), the current falling rate environment is certain to again shine the spotlight on them at regulatory examinations. As such, we thought it would be a good time to review a few key concepts and then go through the best practices to be sure you’re exam ready. So first, the key concepts:

PREPAYMENT ASSUMPTIONS IMPACT BOTH FUTURE EARNINGS AND ASSET VALUES

Prepayment assumptions can have a direct and significant impact on both bank earnings and asset values because they affect the timing and amount of cash flows from loans and mortgage-backed securities (MBSs). Here’s how they impact earnings:

• Interest Income: If prepayments occur faster than expected, loans and MBSs pay off early, typically reducing interest income and requiring prepaid funds to be redeployed, often at lower earning rates.
• Premium Amortization: Financial institutions often buy securities (and sometimes loans) at a premium. Faster prepayments accelerate amortization of that premium, reducing earnings.
• Servicing Income: Prepayments shrink the servicing portfolio, lowering servicing fee income.

Not only do prepayments impact earnings, they also can have a material impact on asset values and the economic value of equity (EVE) calculations (just ask Silicon Valley Bank - R.I.P.). This is because prepayment speeds change the expected duration of assets. That is, faster prepayments shorten duration, making assets less sensitive to rate changes but reducing overall yield and value. Additionally, asset values in EVE simulations are based on discounted cash flow calculations, which use projected cash flows to determine shocked values. If prepayments rise, future cash flows decline, lowering present value. Conversely, if prepayments slow (typically in rising rate environments), durations extend and asset values drop.

THE MARKET RATE ENVIRONMENT AFFECTS PREPAYMENT LEVELS

Prepayment assumptions should vary by rate environment because borrower behavior changes with economic incentives and market conditions. When rates fall, borrowers have a strong incentive to refinance into lower-rate loans, which accelerates prepayments. Lower rates also often stimulate the purchase of homes (and other real estate), increasing turnover as sale proceeds are used to pay off (prepay) loans. Conversely, when rates rise, refinancing becomes unattractive and fewer buyers want to borrow at higher rates to purchase property, so prepayments slow down.

Given these factors, prepayment assumptions on longer-term fixed- and adjustable-rate assets must be dynamic because they are sensitive to interest rate movements and borrower incentives. Static assumptions can lead to inaccurate earnings forecasts and asset valuations.

PREPAYMENT ASSUMPTIONS SHOULD BE INSTITUTION AND CATEGORY SPECIFIC

A “one-size-fits-all” approach typically does not work for asset prepayments. Prepayment assumptions should be category-specific because different loan types and borrower segments exhibit very different prepayment behaviors. Here are some reasons why:

• Fixed-rate loans are more sensitive to rate decreases because borrowers refinance aggressively when rates drop.
• Adjustable-rate loan prepayment patterns depend on reset dates and rate caps, not just market rates.
• Commercial loans often have prepayment penalties, making prepayments less likely.
• High-credit borrowers refinance more quickly when rates fall, whereas lower-credit borrowers may prepay less often due to limited refinancing options.

The bottom line is that a “one-size-fits-all” prepayment assumption can lead to inaccurate forecasts. Category-specific assumptions ensure models reflect actual borrower behavior and product design, improving earnings projections and asset valuation accuracy.

PREPAYMENT BEST PRACTICES

With market rates now declining, it’s likely that loan prepayment assumptions will get much closer scrutiny at regulatory exams. To ensure you pass the test, it’s critical your prepayment assumptions are periodically reviewed, supported, and stress tested.

• All key model assumptions, including prepayments, should be reviewed and updated at least annually (possibly more frequently in times of rapidly changing market rates). Any changes to them should then be reported to, and approved by, the ALCO and full Board of Directors.
• Prepayment assumptions should be institution-specific and be supported by both a quantitative and qualitative analysis. The study should incorporate your portfolio’s actual prepayment history as well as consider any qualitative factors that could impact future levels. Current portfolio durations, along with the rates of the underlying instruments, should be part of that analysis.
• Regulators expect banks and credit unions to stress test (a/k/a “sensitivity testing”) prepayment assumptions by adjusting the regular assumptions to gauge the impact that they have on IRR model results. Best practices are to hold “base case” prepayments constant and then increase falling rate speeds and/or decrease rising rate speeds to simulate a stressed scenario.      

If you need help with your prepayment assumptions, our Advisory Service Team can conduct a prepayment study for you. We also can sensitivity test your assumed prepayment speeds so that you can see the true impact that those assumptions have on your model results. 

Give us a call or email us at advisory@plansmith.com to discuss your organization's individual needs.

Dave Wicklund
Director of ALM Advisory Services