By Patrick Mitchell and Brittany Kleinpaste
ABA DataBank
Discover more in-depth research, dashboards and webinars from the Office of the Chief Economist by exploring ABA’s Economic Research and Insights website.
Executive summary
There is a distinction between how the FDIC determines individual banks’ deposit insurance assessments and how it measures the adequacy of the Deposit Insurance Fund. As a result, the relationship between risk-based pricing, fund size and the reserve ratio is less direct than it was under previous systems.
The FDIC’s reserve ratio is defined as the balance of the DIF divided by estimated insured deposits. The FDIC’s risk-based assessment system uses a different base: average consolidated assets minus average tangible equity (the “assessment base”).
Our analysis (see Figure 1) shows that insured deposits and the assessment base do not always move in lockstep. Because assessment revenue depends on the assessment base while the reserve ratio depends on insured deposits, differences in their growth rates can change the reserve ratio without a corresponding change in assessments, holding assessment rates and other drivers of the DIF constant. For example, starting in 2022, insured deposits grew faster than the assessment base, resulting in slower assessment growth relative to the effect on the reserve ratio.
FDIC Chairman Travis Hill has publicly raised the question of whether insured deposits are the right metric to measure the DIF’s exposure to losses. Chairman Hill noted that an alternative permitted by the FDI Act is to use the assessment base rather than insured deposits as the denominator of the reserve ratio. The FDIC published a staff study in 2007 entitled “An Evaluation of the Denominator of the Reserve Ratio” to consider this very question. The staff study determined that the FDIC considers “it an open question whether at some future time another denominator might be found preferable to estimated insured deposits.” One determination of the study was to wait for additional information and certain reforms to take hold. Since that time, the banking industry experienced a financial crisis with more than 500 bank failures, substantial regulatory reform, and the regional bank failures of 2023, collectively providing a considerably larger body of experience than was available when the staff study was published. Because the reserve ratio determines restoration plans and influences long-run assessment policy, the choice of denominator affects how fund adequacy is measured and when statutory and policy thresholds are reached.
These considerations suggest that evaluating alternative denominator definitions could help clarify the appropriate relationship between fund adequacy metrics and the risk-based pricing framework. This analysis does not assume the assessment base is a superior proxy for losses but highlights a structural disconnect between pricing and adequacy that warrants evaluation of alternative approaches.
Two different concepts: what is being measured vs. what is being priced
The FDIC’s current deposit insurance framework relies on two related but distinct measures:
- Reserve ratio (solvency metric): DIF balance ÷ estimated insured deposits.
- Assessment base (pricing metric): average consolidated assets − average tangible equity. In practice, this is roughly equal to all liabilities, including both insured and uninsured deposits.
These measures serve distinct functions. The reserve ratio guides statutory thresholds and board-designated targets such as the Minimum Reserve Ratio, or MRR, and the Designated Reserve Ratio, or DRR. The assessment base determines the exposure amount to which risk-based assessment rates are applied and therefore plays a central role in translating risk-based pricing into assessment revenue. For example, if the FDIC’s risk-based pricing formula determines a bank should pay five basis points per annum, then the amount the bank would pay would be five basis points multiplied by its assessment base. Conceptually, the assessment base is the exposure measure to which the risk-based rate is applied.
The practical result of the current approach is that the system is funded and individual bank risk is based on one metric (assessment base), while fund adequacy is measured using just one part of that assessment base (insured deposits). For example, a bank with an assessment base (all liabilities) of $10 billion may only have $5 billion of insured deposits as it relies on other forms of funding in addition to insured deposits (uninsured deposits, FHLB advances, trading liabilities, unsecured debt, etc.). So it pays assessments on $10 billion, but only $5 billion of the assessment base is included in the DIF solvency measure. Because the reserve ratio is defined relative to insured deposits, an increase in insured deposits will mechanically reduce the ratio, holding other things constant. If insured deposits grow faster than total liabilities (or vice versa), then the relationship between the two weakens (see Figure 1).
Statutory authority and recent policy discussion
The case for evaluating an assessment-base denominator is not merely conceptual; it is contemplated in statute and has been raised publicly in the context of the DIF restoration plan discussion.
In a May 2025 statement accompanying a restoration plan update, Hill noted that the FDIC moved away from charging assessments on the basis of insured deposits years ago, creating a difference in how assessments are charged and how the health of the DIF is measured. He further stated that an alternative permitted by the FDI Act is to use the assessment base rather than insured deposits as the denominator of the reserve ratio, and that he asked staff to analyze this option for future consideration.
The cited statutory provision (12 U.S.C. §1817(b)(3)(B)) allows the FDIC board to designate a reserve ratio that is not less than 1.35% of estimated insured deposits “or the comparable percentage of the assessment base.” At a minimum, that language supports evaluating how a comparable assessment base threshold would be calibrated and how it would perform relative to the current insured-deposit denominator.
The statutory flexibility reflects the fact that different denominator choices emphasize different concepts of exposure. Insured deposits focus on the portion of liabilities explicitly protected by the DIF, while the assessment base more closely aligns the fund adequacy metric with the base used to determine assessment revenue. By authorizing either approach, Congress provided the FDIC with discretion to evaluate which measure better reflects the relationship among exposure, pricing, and fund adequacy.
Empirical evidence: insured deposits and the assessment base do not always move together
Analysis indicates that insured deposits and the assessment base sometimes grow at similar rates, but they have had different growth dynamics in recent years.
Figure 1 reproduces the internal comparison of cumulative growth rates for the assessment base and insured deposits since the implementation and reporting of the assessment base. The time frames in Figures 1, 2 and 3 begin after the expiration of the Transaction Account Guarantee, or TAG, program to remove fluctuations in insured deposit growth due solely to changes in legislation.
While growth rates of insured deposits and the assessment base were broadly similar through 2022, the growth rates diverged starting in 2022. The observed divergence in 2022 is consistent with, but not solely attributable to, slower asset growth following the unwinding of post-COVID fiscal and monetary policy, along with changes in deposit composition and funding dynamics during the tightening cycle.
The divergence grew in 2023. Insured deposits grew over 3.6% in 2023 while the assessment base slightly declined (-0.6%) over the same time frame. The divergence suggests that insured deposits may react differently than the broader assessment base to changes in funding conditions and monetary policy.
This divergence matters because assessment revenue is generated from the assessment base, while the current reserve ratio depends on insured deposits. When these measures shift differently, the reserve ratio can move mechanically even if the funding base and resulting assessments through risk-based pricing are unchanged or even decline.
Figure 2 compares how the calculated reserve ratios would have performed relative to each other if the translation from insured deposits to assessment base had been calculated in the first quarter of 2013 after the expiration of TAG. This translation is calculated by calibrating the reserve ratio to the assessment base at a common starting point (2013, Q1), the first full quarter after the expiration of the TAG program, to avoid distortions in insured deposit levels. While each of the reserve ratios would have dropped below the MRR following the COVID-related monetary and fiscal policy action, the calculated reserve ratio would have met the MRR sooner using the assessment base as the denominator. This highlights how the two approaches lead to different timing in meeting the minimum reserve ratio.
Figure 3 shows the implied deficit or surplus to the minimum under alternative denominators. This is simply a different way to visualize the performance against a translated MRR using the assessment base versus insured deposits. This analysis converts the percentage above or below the MRR across the two denominators and normalizes it into a comparable dollar amount. For example, in the second quarter of 2025, when the reserve ratio first crossed the MRR using insured deposits, the excess amount was approximately $800 million versus what would have been approximately $4.5 billion using the assessment base as the alternative.
Collectively, these figures illustrate that denominator choice affects not only the level of the reserve ratio but also how fund adequacy is interpreted through time. When insured deposits grow faster than the assessment base, fund adequacy can decline more rapidly than under a broader exposure measure. This disconnect complicates the interpretation of fund adequacy relative to the risk‑based pricing system.
An additional consideration is that the level of insured deposits, and by extension the reserve ratio, can be affected by changes in deposit insurance coverage or depositor behavior. For example, if we wanted to model out the impact of insuring all deposits, rather than capping deposit insurance at $250,000, a stylized full-coverage example illustrates the sensitivity of the current denominator to coverage definitions: the reserve ratio would immediately decline from its current level of 1.43% to roughly 0.81% and would trigger a restoration plan. However, if the FDIC had previously shifted to using the assessment base as the denominator for the MRR in 1Q2013, the MRR would have been established at 0.65%.
In this stylized example of covering all deposits, the MRR would not change as a result of the change in coverage, because insured and uninsured deposits are counted equally in the assessment base, thus avoiding a restoration plan. This example is not intended to evaluate the merits of expanded coverage or to suggest that fund requirements should be unaffected by coverage changes. Rather, it illustrates that the denominator can materially affect how statutory thresholds translate into restoration-plan mechanics.
Risk alignment: why the denominator choice can be economically meaningful
The assessment base and insured deposits capture different aspects of exposure and risk. No single metric fully captures the potential drivers of losses to the DIF. Insured deposits reflect the legally protected obligations of the DIF, even if they do not fully capture realized or expected losses across resolution outcomes.
The denominator question is also linked to how the reserve ratio relates to underlying drivers of failures and losses. Several considerations are relevant when comparing insured deposits and broader balance-sheet measures:
- Whether a deposit is insured is tied closely to depositor behavior in a stress, which can influence failure probability and fund losses in some cases, as evidenced by the three large bank failures in the spring of 2023.
- Historical loss data from the FDIC’s Options for Deposit Insurance Reform report indicate that uninsured depositors avoided losses in 94% of the resolutions over the 2008-22 time frame. This observation suggests that insured deposits alone may not fully capture realized losses to the DIF across resolution outcomes, particularly under resolution strategies that protect uninsured depositors. At the same time, loss outcomes depend heavily on resolution strategy and failure conditions, which weakens the case for any single liability-based metric fully capturing DIF exposure.
- The regional bank failures in 2023 highlighted the importance of broader balance-sheet characteristics, including liability composition, in informing loss outcomes to the DIF.
- For example, the FDIC estimates the special assessment for the failures of Silicon Valley Bank and Signature Bank to collectively be approximately $16.7 billion as of April 17, 2026. This estimate of loss is attributable solely to losses associated with coverage of uninsured depositors. Silicon Valley Bank had 88% uninsured deposits at failure and Signature had 67% uninsured deposits at failure.
- Secured bank funding at failure is repaid from pledged collateral and therefore reduces the pool of unencumbered assets available to satisfy unsecured claims, including depositors. A measure that considers only insured deposits does not account for the risk of these senior liabilities.
- The relationship between insured deposit share, franchise value at failure, and resolution costs is complex and does not follow a consistent pattern.
Next steps
Nearly two decades have passed since the FDIC last published a comprehensive study of the denominator question, during which the banking system has experienced a financial crisis, major regulatory reform, and the 2023 regional bank failures. The FDIC should undertake and publish a comprehensive reassessment of the reserve-ratio denominator using the expanded body of post-2007 evidence. Together these developments provide a richer evidence base than existed when the FDIC last examined the issue.
A comprehensive reassessment should include a comparative analysis of alternative denominator definitions. The analysis should compare reserve-ratio performance under alternative denominators, evaluate their relationship to historical loss outcomes and assessment revenue, and assess their sensitivity to changes in deposit composition and macroeconomic conditions. Such analysis would not presuppose changes in the existing framework but would clarify how alternative metrics compare in representing exposure, supporting the risk-based pricing framework, and measuring fund adequacy.
Whether insured deposits remain the preferred denominator or the assessment base proves more informative is ultimately an empirical question. Nearly two decades after the FDIC last examined the issue, the empirical foundation for answering that question is substantially stronger.
Patrick Mitchell is head of economic policy research at ABA. He previously served as director of the Federal Deposit Insurance Corporation’s Division of Insurance and Research. Brittany Kleinpaste is a VP for economic research at ABA. The views expressed are those of the authors and should not be attributed to the Federal Deposit Insurance Corporation.
