This essay is based on the paper “Interest Rate Risk in Banking” by Peter DeMarzo, Arvind Krishnamurthy, and Stefan Nagel.
The dramatic collapse of Silicon Valley Bank (SVB) in March 2023—and the broader turmoil in the US banking sector that followed—raised pressing questions for policymakers, regulators, and the public. Were the bank failures we observed a result of a temporary liquidity shortfall, or a sign of deeper insolvency? How pervasive were these risks at other regional banks? And how can we tell when a bank should be liquidated versus restructured?
In a new study, we provide a framework to evaluate the long-term health of banks by focusing on its franchise value—the expected future profits a bank earns from its core business of lending and deposit-taking. Our analysis offers a way to assess whether banks are fundamentally solvent, even when market shocks temporarily depress the value of their assets.
What Went Wrong in 2023?
The core of the 2023 crisis was interest rate risk. In the years leading up to the crisis, banks like SVB invested heavily in long-term, fixed-rate securities while funding themselves primarily with short-term deposits. When the Federal Reserve rapidly raised interest rates, the market value of those long-term assets fell. Some banks found that the value of their assets had dropped below their liabilities—raising alarms about insolvency.
But this mark-to-market snapshot is not the entire story. A bank that would be insolvent if closed today may still be a viable business. If it has a strong customer base and can continue to generate profits from its lending and deposit operations, it may be able to recover the current shortfall over time. This is where franchise value comes in.
The Role of Franchise Value
Franchise value is the present value of a bank’s expected future profits from offering financial services. It includes the spreads banks earn on both deposits and loans—essentially, the difference between what they pay depositors and what they earn from borrowers—minus operating costs like salaries and rent.
This value can be large. Even if a bank’s investment portfolio suffers losses, a healthy stream of future profits can help it remain solvent over the long run. That’s why understanding franchise value is critical for regulators deciding whether to intervene during periods of stress.
The Myth of the Negative Duration Hedge
A popular belief—reflected in both academic research and regulatory guidance—is that banks with “sticky” deposits (where deposit rates rise less, and more slowly, than market rates) actually benefit when interest rates rise. The logic is that rising rates increase the spread between what banks earn and what they pay out, supposedly boosting franchise value. Under this logic, the losses on the securities held by banks such as SVB were offset, or hedged, by the increase in franchise value.
We show that this belief is flawed: While banks with low “deposit betas”—meaning few deposits leave these banks when market interest rates rise—do indeed earn higher spreads as rates rise, the value of these future earnings doesn’t necessarily go up. In fact, when evaluated properly using standard valuation methods, the present value of future profits from deposit spreads is relatively independent of the level of interest rates (i.e., has zero duration). At the same time, the present value of the banks’ lending spreads, which are largely fixed, tend to fall when rates rise. That means the franchise value of the median US bank actually declined as interest rates increased in 2022–23—adding to, rather than hedging, losses from falling asset prices.
A Better Way to Measure Risk
To evaluate franchise value, we construct a model that translates a bank’s loan and deposit activities into a set of replicating financial instruments—such as floating-rate and fixed-rate bonds. This allows us to estimate the duration (interest rate sensitivity) of the bank’s franchise value.
Our key finding is that, for a large majority of US banks, franchise value has positive duration. This means that as interest rates rise, the present value of the bank’s future profits declines. This is in contrast to prevailing regulatory guidance, which recommends treating banks’ deposit bases as long-duration fixed liabilities that would lead to a gain in bank value when rates rise.
By applying our model to a large dataset of US banks, we find that the median bank’s franchise value declined by about 2.2% of assets during the recent interest rate hikes—compounding losses from long-term securities holdings, which fell by roughly 3.6% of assets.
Solvency and Stability
Despite these losses, our analysis suggests that most US banks remain solvent when franchise value is properly accounted for. In particular, the banks that experienced the largest mark-to-market losses on securities in 2023 also tended to have the highest franchise values—providing a buffer against insolvency.
Importantly, our findings do not suggest that all is well in the banking system. Banks have increasingly sought to hedge income volatility (cash flow risk) by taking on duration risk—investing in long-term securities to stabilize net interest margins. This may help short-term earnings appear stable but exposes banks to larger swings in market value, making them more fragile in periods of rising rates.
Policy Implications
Our results carry several key implications for policymakers:
- Franchise value matters. Solvency assessments must go beyond balance sheet snapshots. A forward-looking evaluation of a bank’s earnings potential is essential to determining whether it should be supported during a crisis.
- Misguided regulation can increase risk. Current regulatory frameworks that mischaracterize the duration of deposit franchises may encourage banks to hold more long-duration assets, inadvertently increasing systemic risk.
- Forbearance is sometimes justified—but not always. If a bank’s franchise value is strong, regulators may be right to support it during temporary stress. But they must also guard against the emergence of “zombie banks” that survive only by gambling with taxpayer backstops.
- Business models are not static. Our model assumes banks continue to earn profits in the future much as they have in the past. But changes in depositor behavior—such as shifts to higher-yield online accounts—or competition from nonbank lenders could undermine this assumption.
A Caution and a Call to Action
The failures of SVB and others were not isolated incidents. They reflect a broader misunderstanding of how banks create value and how that value changes when interest rates rise. Our framework provides a clearer lens for assessing solvency—and a warning that today’s regulatory tools may be built on outdated assumptions.
To ensure financial stability, we must rethink how we measure and manage interest rate risk in the banking system. That starts with better models—and better policies—grounded in the realities of bank franchise value.
Read the full paper here.
Peter DeMarzo is the John G. McDonald Professor of Finance at the Stanford Graduate School of Business.
This essay is part of the Financial Regulation Research Brief Series. Research briefs in this series highlight the policy-relevant research on financial systems, including the impact of financial regulations on economic growth and stability.
