Liquidity Risk vs. Interest Rate Risk
Liquidity risk is fundamentally different from the interest rate risk that dominates most ALM discussions—and this distinction is critical for survival in a crisis. Interest rate risk is about the timing and magnitude of your cash flows in a stable funding environment. Liquidity risk is about whether you can meet obligations at all when funding markets dry up.
When the Federal Reserve raised rates in 2022–2023, Silicon Valley Bank had locked in a massive book of long-duration Treasury securities yielding 1–2%. Interest rate risk was severe: the economic value of their equity plummeted. But that alone wouldn't have killed SVB. What killed SVB was liquidity risk—uninsured depositors realized the bank had a major mark-to-market loss and created a run. Within 48 hours, $42 billion of deposits fled. SVB couldn't sell $91 billion of underwater securities without crystallizing losses that would exceed capital. That's liquidity risk.
Liquidity risk asks: "Can we raise cash when we need it, at what cost, and under what market conditions?" It's about access to funding, not earnings sensitivity.
Why This Matters
For a CIO or Treasurer, interest rate risk is managed by matching duration and convexity. You hedge with futures, swaps, or caps. Liquidity risk requires a different mindset: assume your funding sources can disappear overnight. Regulators now enforce this assumption through the Liquidity Coverage Ratio (LCR) and Net Stable Funding Ratio (NSFR), stress tests, and contingency funding plans.
The 2023 regional bank crisis crystallized this. Banks with strong capital ratios and reasonable interest rate risk positions still failed because they couldn't answer a simple question: "If depositors and wholesale creditors get nervous tomorrow, do we have enough liquid assets to survive 30 days (LCR) or 1 year (NSFR)?"
Core Concepts
Funding Stability: Which deposit and wholesale funding sources are sticky? Uninsured deposits are volatile; insured deposits are stable. Term wholesale funding is more stable than overnight repo. You must model deposit outflows under stress—both idiosyncratic (bank-specific) stress and systemic stress.
High-Quality Liquid Assets (HQLA): Not all assets are equally liquid. A Treasury bond is liquid; a residential mortgage takes weeks to sell. Regulators define HQLA strictly: US Treasuries, agency MBS, and select other securities get haircuts ranging from 0% to 50%. Your HQLA buffer is your lifeline in a crisis.
Net Liquidity Position: This is cash inflows minus outflows over a stress scenario. If you're $5 billion net negative over 30 days, you need $5 billion of HQLA to pass LCR. If you're $20 billion net negative over 1 year, you need to fund that gap with "stable" funding sources or HQLA.
Stress Assumptions: Regulatory stress scenarios differ from market stress scenarios. Regulators assume 5% deposit outflow for insured deposits, 25% for uninsured, and sometimes 100% for brokered deposits. Real stress can be worse: in March 2023, Signature Bank and SVB faced 100%+ withdrawal rates for uninsured deposits.
How It Connects to Balance Sheet Management
Liquidity risk is integrated with interest rate and credit risk. A bank that funds long-term loans with short-term deposits is exposed to both interest rate risk (if rates rise, deposit costs increase faster than loan yields) and liquidity risk (depositors may flee before rates stabilize). A bank that holds illiquid assets (illiquid mortgages, commercial real estate loans) funded by volatile funding sources faces acute liquidity risk.
Your balance sheet structure—the mix of assets and liabilities—is the root cause of liquidity risk. ALM professionals must think beyond duration and convexity to ask: "If a stress event occurs, can we sustain this balance sheet?"
The strongest ALM frameworks integrate three components:
1. Structural liquidity management: Design funding sources and asset holdings so you can weather stress
2. Dynamic liquidity management: Monitor and adjust in real-time (internal stress testing, contingency plans)
3. Regulatory compliance: Meet LCR, NSFR, and other regulatory standards under defined stress scenarios
All three require understanding liquidity risk as distinct from—and often more critical than—interest rate risk.
The Strategic and Tactical Divide: Why These Risks Are Fundamentally Different
The first principle experienced ALM practitioners must internalize is deceptively simple: liquidity risk and interest rate risk are not different points on the same spectrum. They are different dimensions altogether, each requiring distinct governance, measurement, and mitigation strategies. Understanding why they differ—and how they interact—is foundational to effective balance sheet management.
Interest rate risk operates within a framework of continuous change. Your earnings sensitivity to a 100 basis point rate move is measurable, forecastable, and tradeable. If you deploy a Treasury hedge today to reduce that sensitivity, you've locked in protection. The fundamental economics don't shift simply because you acted. Interest rate dynamics follow relatively predictable paths: rates move gradually (in expectation), markets for hedging instruments remain liquid, and you can price the cost of protection with precision.
Liquidity risk operates in an entirely different domain—one of discrete shocks and binary outcomes. It's the difference between "rates might rise" and "we can't fund ourselves at any price." A $2 billion wholesale funding maturity next Tuesday is a tactical problem that requires precise cash flow management. But a customer relationship managing $500 million in insured deposits represents strategic stability: those deposits reliably remain because they're protected by FDIC insurance and the customer has limited incentive to move them. The same $500 million, if moved into brokered deposits intermediated through a third party, suddenly becomes tactical—it can redeem on any given day with virtually no notice, because it's held on a platform where depositors can switch to competitors on a mouse click.
Interest Rate Risk: The Maturity Gap Framework
Classic interest rate ALM is built on the concept of the maturity (or repricing) gap: the fundamental mismatch between how quickly your assets and liabilities reprice when market rates change. Consider a bank with $100 billion of fixed-rate residential mortgages funded by $60 billion of variable-rate deposits. This creates a positive repricing gap—if interest rates rise, the bank's deposit costs increase immediately, but mortgage yields are locked in. The net interest margin contracts as the funding rate rises toward the asset yield.
The remedy is structural and mechanical: a pay-fixed interest rate swap, a rate cap, or a shift in funding sources. The calculation is precise because interest rates follow predictable patterns. You can model the net interest income impact on a spreadsheet using standard IRR curves, stress it under multiple rate scenarios, and quantify exactly how much protection a hedge provides. A 100 basis point rate move has a calculable impact on net interest margin and on the economic value of equity. This predictability is why interest rate risk frameworks rely on sensitivity analysis: "If rates rise 1%, our NII declines $X." It's measurable, continuous, and tradeable.
Liquidity Risk: The Funding Shock Framework
Liquidity risk doesn't work that way. It's not about "slightly higher funding costs." It's about funding disappearing entirely. When a discrete event occurs—a peer bank fails, credit spreads widen 200 basis points in a matter of hours, a systemically important counterparty defaults, or depositors lose confidence—funding doesn't become "harder to get." It becomes unobtainable.
Signature Bank in March 2023 is the quintessential example. In a 24-hour period, the bank faced uninsured deposit outflows exceeding 100% of available uninsured deposits. There was no pricing mechanism that would have kept those deposits in place. No matter what rate the bank offered, sophisticated depositors with access to insured alternatives (Treasury money market funds, stable value funds, other banks) were walking out the door because the perceived risk of loss exceeded the offered return.
This is why liquidity frameworks are built on scenario analysis rather than sensitivity analysis. You don't model "liquidity sensitivity to a 0.5% deposit outflow." Instead, you construct discrete stress scenarios: "What if we lose 25% of uninsured deposits in a single week AND wholesale funding spreads widen 200 basis points AND rating agencies downgrade us?" Each scenario is a specific shock; the outcomes are often binary—the bank either makes it through, or it doesn't.
The 2008 and 2023 Lessons: When the Liquidity Constraint Becomes Binding
The 2008 financial crisis taught regulators an essential lesson: a well-capitalized bank with positive net worth can still fail if it cannot access funding. Lehman Brothers had capital. It had assets. It had the ability to generate earnings. What it didn't have was access to liquidity when it needed it, because wholesale funding markets seized up completely. Counterparties wouldn't finance Lehman paper at any price.
This realization prompted a fundamental shift in regulatory philosophy. Prior to 2008, capital requirements were the binding constraint on bank safety. Post-2008, the Dodd-Frank Act, Basel III, and subsequent prudential rules introduced explicit liquidity requirements—the LCR (Liquidity Coverage Ratio) and the NSFR (Net Stable Funding Ratio)—that treat liquidity as an independent binding constraint. A bank could theoretically have adequate capital but insufficient liquidity and still face regulatory pressure or failure.
SVB and the regional bank stress of March 2023 reinforced this lesson powerfully. SVB had capital ratios above regulatory minimums. On a purely mechanical level, it was solvent. But it had two fatal vulnerabilities: (1) a highly concentrated deposit base (over 70% uninsured) sourced primarily from a single customer segment (venture-capital-backed technology firms), and (2) a portfolio of illiquid assets (residential mortgage-backed securities) with steep mark-to-market losses when rates rose.
When depositors lost confidence in tech sector funding (a realistic concern in early 2023), they had every incentive to withdraw. SVB couldn't hold them with rate increases because many had alternative homes for their cash (government-backed deposits, Treasury money market funds, other banks). The bank couldn't raise liquidity by selling assets because MBS prices had declined sharply, and selling $10 billion of MBS into a wide-bid-ask market would destroy value. Within 48 hours, the bank faced uninsured deposit outflows that exceeded its liquidity buffer.
First Republic, Signature Bank, and other failures that followed all shared this pattern: concentrated, sophisticated deposit bases (uninsured) funding illiquid asset portfolios, with insufficient standalone liquidity buffers and over-reliance on wholesale funding that dried up instantly when confidence evaporated.
Why Interaction Between the Two Risks Matters Critically
Although interest rate risk and liquidity risk are distinct, they interact in ways that amplify vulnerability. When you fund long-term fixed-rate assets (mortgages, long-duration corporate bonds) with short-term deposits, you create both risks simultaneously.
From an interest rate perspective: If rates rise, your net interest margin compresses because deposits reprice up faster than mortgages. Your earnings suffer.
From a liquidity perspective: If depositors get nervous about any reason—economic slowdown, competitor offering higher rates, loss of confidence in the bank's capital position—they can withdraw their deposits at any time. You're forced to either raise rates dramatically (destroying profitability) or sell illiquid assets at market prices (crystallizing losses).
SVB's situation embodied this interaction perfectly. The bank had purchased $120 billion of long-duration MBS and government securities at yields of 1.5%–2.5% during the ultra-low-rate environment of 2020–2021. When the Fed raised rates to 4.25%–4.5% by late 2022, two things happened:
1. Interest rate risk materialized: The economic value of those securities declined by roughly 15%–20%. On a mark-to-market basis, the bank's securities portfolio was under water by ~$15 billion.
2. Liquidity risk materialized: Depositors realized the bank's capital cushion had eroded. Combined with the broader tech sector slowdown, they rationally decided to withdraw. The bank couldn't fund the withdrawals because (a) it couldn't sell securities without crystallizing losses that would wipe out remaining capital, and (b) wholesale funding markets simply closed.
A bank that had avoided this interaction—by maintaining a shorter-duration asset portfolio, funding with more stable sources, and keeping a larger HQLA buffer—would have survived 2023 intact.
The Strategic Implication: Liquidity as the Binding Constraint
For most banks, liquidity risk is the more dangerous of the two. A bank can tolerate modest interest rate risk because it has tools (hedges, balance sheet management, repricing) to manage it continuously. But liquidity risk is binary: if you don't have it when you need it, hedges are irrelevant. Capital is irrelevant. You're insolvent by definition.
This suggests a practical priority: if you're managing a $100 billion bank and can allocate 80% of ALM resources to one risk and 20% to another, the math strongly favors liquidity. Yet many banks historically inverted this priority, focusing ALM effort on interest rate hedging and treating liquidity as a compliance checkbox.
The strongest banks in 2023 had simple deposit bases (less reliance on uninsured, wholesale, or brokered funding), robust HQLA buffers (30%+ of liabilities in liquid securities), and disciplined wholesale funding plans (laddered maturities, diversified sources). They had less interest rate risk hedging because they didn't need it—their balance sheet structure was naturally conservative. SVB, Signature Bank, and First Republic optimized aggressively for profitability in a low-rate environment and paid the price when liquidity became scarce.
Practical Framework for Integrated Monitoring
A best-practice ALM function monitors these risks separately but with explicit attention to interaction:
Interest Rate Risk:
- Measure net interest income sensitivity under Fed rate scenarios (100 bp up/down)
- Measure economic value of equity under parallel rate shifts
- Calculate earnings at risk (EaR) over a 12-month horizon
- Monitor duration gap (faster repricing assets minus faster repricing liabilities)
- Govern through target ranges (e.g., "NII should not decline >5% under a 100 bp up scenario")
Liquidity Risk:
- Measure LCR and NSFR under regulatory stress scenarios
- Forecast daily, weekly, and monthly cash position under normal and stress conditions
- Stress deposit outflows by segment (insured vs. uninsured, customer type, geography)
- Measure HQLA coverage (total liquid assets as % of expected outflows)
- Govern through target ranges (e.g., "LCR > 120%, NSFR > 110%, uninsured deposits < 25% of total deposits")
Interaction Monitoring:
- Recognize that duration risk (long-duration assets) + funding risk (volatile deposits) creates a combination vulnerability
- In quarterly ALM reviews, explicitly discuss: "Our mortgage book is now $X of duration; if rates move 200 bps, we have mark-to-market losses of $Y. Our uninsured deposits are $Z. Can we absorb this scenario without forced asset sales?"
- Build buffers that account for both risks simultaneously: adequate HQLA, not just for the LCR scenario, but also for a scenario where mark-to-market losses on the securities portfolio erode capital
- Stress test: "If we lose 50% of uninsured deposits in 10 days AND the Fed funds rate rises 100 bps (creating a $2 billion mark-to-market loss), can we fund operations?"
This integrated mindset is what separates banks that weathered 2023 from those that didn't.