Are Your Deposit Assumptions Grounded in Reality?
- Michael Hunker
- 2 days ago
- 4 min read
For many years, non-maturity deposit modeling occupied a relatively stable place in balance sheet analytics. Assumptions were established, reviewed periodically, and often left unchanged for long stretches of time. Core deposits were viewed as reliable funding, and modeling frameworks reflected that confidence.
The most recent interest rate cycle disrupted that stability. Rates rose rapidly. Deposit pricing competition intensified. Balances shifted. Institutions that had relied on static deposit assumptions suddenly faced outcomes that diverged from expectations and dynamic modeling increasingly became a best practice.
What the industry learned was not that deposit behavior changed; it was that deposit behavior has always been dynamic. The recent cycle simply made that dynamic nature impossible to ignore.
Static Assumptions in a Dynamic Environment
Static assumptions simplify reality. A single decay rate or a single beta applied across products and rate environments can appear stable and conservative. During calm periods, that stability can feel appropriate.
But depositor behavior responds to incentives. When market rates rise, the return available on alternative investments increases, making low-yielding deposits less attractive. When rates fall, that pressure eases and pricing sensitivity declines. When spreads narrow, retention improves. As accounts season, stability often increases.
If assumptions remain fixed while these forces shift, models gradually lose alignment with observable behavior.
What Do We Mean by “Decay”?
Decay is not an abstract academic construct or a one-time study output. Rather, it’s a measure of how deposit balances behave over time.
When balances decline, that is decay. When existing balances grow, realized decay can be lower, or even negative. These movements reflect customer decisions in response to pricing, alternatives, and economic conditions.
Because decay represents observable behavior, it can be measured, trended, and back-tested against actual account performance.
For that reason, deposit assumptions in a model should not be arbitrary. They should be grounded in empirical evidence and periodically evaluated against realized results.
What Are Some Variables that Drive Deposit Behavior?
Across multiple rate cycles and billions of account records, four factors consistently explain behavior patterns:
1. Account Age
New balances tend to decay at higher rates. As accounts season, they stabilize. Treating all balances as equally seasoned ignores one of the strongest predictors of deposit behavior.
2. Account Size
Large balances are typically more rate sensitive and more actively managed. Smaller balances often behave transactionally and may even grow. Decay patterns frequently reflect a reversion toward typical balance levels.
3. Pricing Relative to Competitors
Depositors evaluate rates in context. Accounts priced below competitive alternatives experience higher runoff. Competitive pricing improves retention. Static assumptions rarely capture this relative dynamic.
4. Prevailing Market Rates
The broader rate environment influences opportunity cost. As market rates rise, deposit pricing betas tend to increase and effective duration declines. Academic research, including work from the Federal Reserve Bank of Dallas, has documented this dynamic behavior.
Together, these drivers explain why deposit behavior is not constant across time or environments. Additional factors such as seasonality, product age, and burnout also influence outcomes.
Governance, Regulators, and the Real Objective
Regulators do not prescribe specific assumption levels, but they expect assumptions to be reasonable, supported, and well-documented. Movement in assumptions is not inherently problematic, but unexplained movement is, which is why back-testing and clear rationale are essential.
It is also important to distinguish between modeling objectives. Static assumptions can still serve a role in policy compliance and reporting, where consistency is required. However, for strategic analysis and understanding earnings behavior in changing environments, assumptions that reflect current conditions are more informative.
The objective is not to eliminate variability, but to ensure assumptions are grounded in observable data, economic logic, and common sense. Deposit modeling should not feel like a black box. Assumptions should be explainable, defensible, and clearly connected to how customers actually behave.
What This Means in Practice
As you review your own deposit assumptions, consider a few practical questions:
When was the last time you validated your assumptions against actual account behavior?
Do your assumptions reflect current pricing dynamics and rate conditions, or long-term averages?
Can you clearly explain how you derived those assumptions and why they are still appropriate today?
Deposit behavior has always been dynamic; today’s environment has simply made that reality more visible. When institutions rely on static assumptions, they risk falling out of sync with how customers actually respond to changing rates, pricing, and market conditions. But by grounding assumptions in observable data, regularly validating them against performance, and incorporating key behavioral drivers, institutions can help ensure their models reflect reality and perform as intended.
If you have questions about deposit assumptions, contact the DCG team to discuss your approach and help ensure your modeling framework aligns with today’s environment.
For more insights from Darling Consulting Group, click here.
Michael Hunker is a Solutions Consultant at Darling Consulting Group. He has a comprehensive background in risk management, financial analysis, and data analytics. In his role, he leverages his expertise to help financial institutions optimize their deposit strategies and manage risks effectively. He presents sophisticated deposit model results to senior executives and provides in-depth training on DCG's software solutions.
Before joining DCG, Michael was an Asset Liability Manager at First Tech Federal Credit Union, where he played a pivotal role in implementing an ALM modeling platform and managing a substantial investment portfolio. His responsibilities included presenting critical liquidity and interest rate risk analyses to the executive team and ensuring compliance with regulatory standards.
Michael holds a Bachelor of Science in Accounting and a Master of Science in Finance from Pacific University, and he is a Certified Treasury Professional (CTP).
© 2026 Darling Consulting Group, Inc.