Deposits are US banks’ primary funding source; they are typically cheap, abundant, and slow to reprice. Many households keep substantial balances in bank accounts that pay little or no interest. When the Fed hikes the policy rate, banks only pass part — if any — of the increase to deposit rates. But deposits can also add to banks’ interest rate risk, as Silicon Valley Bank’s failure in March 2023 made painfully clear. For bank supervision, a key question is therefore when deposits hedge banks’ interest rate risk and when they exacerbate it.

One useful starting point is the deposit franchise value, i.e. the present value of the profits banks earn from paying less than the policy rate on their deposits. How the deposit franchise value changes with the policy rate captures the interest rate risk in deposit funding. If the deposit franchise value rises when the policy rate rises, i.e. negative duration, deposits hedge interest rate risk; otherwise, they exacerbate it. A growing and influential literature connects DFV duration to a bank’s deposit beta, which measures the pass-through of policy-rate changes to deposit rates. One view — associated with the low-deposit beta evidence highlighted by Drechsler et al. (2017, 2021) and Drechsler et al. (2025) — emphasises that deposits can have negative duration, especially with fixed operating costs. Another — articulated by DeMarzo et al. (2025) — cautions that low deposit betas are not, by themselves, evidence of negative deposit franchise value duration. They highlight that banks’ franchise value (loan franchise value plus deposit franchise value) has positive duration, especially when banks are expanding. These discussions point to an uncomfortable takeaway for the post-SVB debate: deposit beta may not be a sufficient statistic for bank deposits’ interest rate risk, and the hedge of deposits can fail precisely when it is most needed. 

A key premise behind the findings of low deposit beta is that deposits are sticky: depositors do not withdraw much when rates move. Which depositors are sticky, and why? A new wave of research is now opening this black box with micro evidence on depositor heterogeneity. Lu et al. (2025) show that payment delays dampen US depositors’ responses to interest rates. Argyle et al. (2025) find that stickiness is disproportionately driven by a small set of high-balance accounts that are particularly insensitive to rate changes. Cirelli and Ólafsson (2025) document widespread household inaction among Icelandic depositors. Blickle et al. (2025) show that depositor rate sensitivity varies over time and matters for aggregate financial stability. Egan et al. (2025) formalise bank competition with sleepy deposits, motivated by the fact that depositors rarely switch banks. Benetton et al. (2026) further show that depositors who switch in are systematically different from a bank’s existing customers, highlighting that the existing depositor base is a valuable asset for banks.

In Lu and Wu (2025), we advance both the deposit franchise value duration discussion and the literature on depositor heterogeneity by documenting new micro evidence of depositor inattention and showing how it affects both banks’ deposit pricing and interest rate risk. Inattention is pervasive in the contexts of macroeconomics and household finance (e.g. Coibion and Gorodnichenko 2012, Pagel and Ólafsson 2018). Deposit markets are naturally prone to inattention: monitoring rates is time-consuming, and the private gains are often small at the individual level. Consistent with this, we show that depositors manage their accounts infrequently. Specifically, we track how individuals react to income from different sources. Some income payments are scheduled with known payment dates (such as regular salary income), while others arrive on unknown dates (such as tax refunds or fiscal checks). If depositors manage their bank accounts continuously, they should react at a similar speed to income from different sources. Empirically, account balance adjustments following tax refunds are much slower than adjustments to regular income and vary across depositors. Moreover, the depositors who are slow to react to unscheduled inflows are also less responsive to interest rate changes around FOMC announcements: after policy rate hikes, they are less likely to withdraw.

Figure 1 Banks with more inattentive depositors offer lower rates

Notes: In each year, US banks are ranked by their deposit-weighted inattention measure and grouped into quartiles. For each quartile, we compute the average savings deposit rate as interest expense on savings deposits divided by savings deposits using US Call Reports, 2007-2021. The figure plots four-quarter moving averages for the top quartile (more inattentive depositors; blue dashed line) and the bottom quartile (more attentive depositors; orange solid line).

We interpret these patterns as evidence of depositor inattention and use it as the organising lens to study its banking implications. Constructing and aggregating depositor inattention based on the reaction time gaps between scheduled and unscheduled income, we find that inattention is informative at the bank level: banks serving more inattentive depositors set lower deposit rates (Figure 1), and have lower deposit betas and weaker outflows when the policy rate rises.

Motivated by these facts, we propose an analytical model of optimal rate setting featuring depositor inattention. Banks trade off current profit margin (i.e. the spread between the policy rate and the deposit rate) against spread-sensitive outflows from inattentive depositors. If banks set lower deposit rates today, they earn higher profits immediately but will have a smaller depositor base moving forward. Given that the deposit flow sensitivity is modulated by inattention, if depositors are more inattentive, banks set a lower deposit rate. Moreover, when the Fed hikes the policy rate, banks value the future profits less, leading them to set a larger deposit spread today, i.e. passing less of the rate increase to their depositors. 

Figure 2 Deposit franchise value (DFV) duration varies with the policy rate

Note: The figure plots the model-implied duration of deposit franchise value (DFV) per dollar of deposits against the policy rate. At low rates (red solid curve), banks keep deposit rates at zero and DFV rises with rates (negative duration). Beyond that point, DFV falls as rates rise, i.e., positive duration (blue dashed curve). At sufficiently high rates (black dash-dot curve), banks begin paying positive deposit rates. See Lu and Wu (2025) for full details.

While deposit franchise value duration has important policy implications, it is difficult to estimate empirically given the limited time-series variation in the policy rate. Accounting for banks’ optimal rate setting in the presence of depositor inattention, our model suggests that the deposit franchise value duration is state-dependent (Figure 2). It is negative under a low policy rate and turns positive as the policy rate continues to rise. When the level of the policy rate is low, banks optimally keep deposit rates at zero. In this case, a marginal rate hike increases deposit franchise value because the increase in banks’ profit margin dominates the acceleration of deposit outflows and the higher discounting of future profits, leading to negative duration of deposits. When the policy rate is high enough, banks optimally pay positive deposit rates to retain depositors, trading off current margins against future outflows. At the optimum, the marginal gain in spread is exactly offset by faster deposit decay (by the envelope theorem). In this case, a further increase in the policy rate works solely through discounting: it raises the discount rate applied to future deposit profits, thus reducing deposit franchise value as the rate rises. As a result, deposit franchise here has positive duration.  

We refer to this as a model of banks’ temporal monopoly to emphasise the intertemporal trade-off between the profits today and in the future. This trade-off hinges on how banks discount tomorrow’s profits. Discounting matters in any pricing decisions, but it is especially stark for deposit pricing because the policy rate plays a dual role here: as the interest rate on deposits, it sets today’s profit margin for banks; and as the discount rate, it determines how much banks value future profits. This discounting effect is stronger with deposit inflows, in which case future profits account for a larger share of the deposit franchise value.

This model also clarifies how digital banking can reshape deposit markets. In a recent VoxEU column, Acharya et al. (2024) discuss whether it is time to reassess deposit stickiness in a world of digital banking and rapid information flows. In our framework, bank comparison tools and open-banking reforms that lower the cost of monitoring rates and switching banks would increase attention and shrink the deposit franchise. Our calibration suggests deposit franchise value is economically large and sensitive to attention: a 10% increase in depositor attention reduces the deposit franchise value by about 6.6%, with especially pronounced effects when interest rates are already high.

In conclusion, deposit franchise has been and will likely remain central to banking, but whether it cushions or amplifies banks’ interest rate risk – and to what extent – depends on the level of the policy rate and on depositor inattention, two factors that can change quickly with technology and the monetary cycle.

References

Acharya, V V, E Carletti, F Restoy and X Vives (2024), “Banking turmoil and regulatory reform”, VoxEU.org, 7 June.

Argyle, B, B C Iverson, J D Kotter, T Nadauld and C Palmer (2025), “The Dynamics of Retail Deposit Balances”, NBER Working Paper No. 34742.

Benetton, M, B Hébert and T McQuade (2026), “Deposit Competition and Pass-Through: Theory and Evidence Beyond Rates”, mimeo.

Blickle, K, J Li, X Lu and Y Ma (2025), “The Dynamics of Deposit Flightiness and its Impact on Financial Stability”, NBER Working Paper No. 34128.

Coibion, O and Y Gorodnichenko (2012), “Paying attention to inattention”, VoxEU.org, 24 August.

Cirelli, F and A Ólafsson (2025), “What Makes Depositors Tick? Bank Data Insights into Households’ Liquid Asset Allocation”, CEPR Discussion Paper No. 20612.

DeMarzo, P, A Krishnamurthy and S Nagel (2025), “Interest Rate Risk in Banking”, NBER Working Paper No. 33308.

Drechsler, I, A Savov and P Schnabl (2017), “The Deposits Channel of Monetary Policy”, Quarterly Journal of Economics 132(4): 1819–1876.

Drechsler, I, A Savov and P Schnabl (2021), “Banking on Deposits: Maturity Transformation without Interest Rate Risk”, Journal of Finance 76(3): 1091–1143.

Drechsler, I, A Savov, P Schnabl and O Wang (2025), “Deposit Franchise Runs”, NBER Working Paper No. 31138.

Egan, M L, A Hortaçsu, N A Kaplan, A Sunderam and V Yao (2025), “Dynamic Competition for Sleepy Deposits”, NBER Working Paper No. 34267.

Lu, X and L Wu (2025), “Banking on Inattention”, NBER Working Paper No. 34783.

Lu, X, Y Song and Y Zeng (2025), “Tracing the Impact of Payment Convenience on Deposits: Evidence from Depositor Activeness”, Jacobs Levy Equity Management Center for Quantitative Financial Research Paper, The Wharton School Research Paper.

Pagel, M and A Ólafsson (2018), “The ostrich in us: Selective attention to personal finances”, VoxEU.org, 7 August.