Skip to main content

Rate Sensitivity

Updated Jun 24, 2026 at 8:22pm

Research Draft Medium 1,270 words

Rate sensitivity is the degree to which a bank's earnings, and ultimately its stock, respond to changes in interest rates. A bank is fundamentally a maturity-transformation machine: it funds itself with deposits and short-term borrowings and lends or invests at longer maturities. Whether rising or falling rates help or hurt depends on which side of the balance sheet reprices faster. The core tension is that the same rate move can expand one bank's net interest margin while crushing another's — and the asset side that looks safe on a credit basis (Treasuries, fixed-rate mortgages) can carry the largest interest-rate risk, as the 2023 regional-bank failures demonstrated.

How it's measured

Two complementary lenses dominate bank rate-risk analysis (FDIC, Sensitivity to Market Risk examination manual):

  • Net Interest Income (NII) sensitivity — an earnings view. Banks run simulations of NII under parallel rate shocks (e.g. +/-100, +/-200, +/-300 bps over 12 months) and report the percentage change. A positive NII change under a rate rise means the bank is asset-sensitive; a negative change means liability-sensitive.
  • Economic Value of Equity (EVE) — a long-horizon view. EVE marks all assets and liabilities to present value and measures how the net position changes under a rate shock. It captures the duration risk that a 12-month NII simulation misses — exactly the risk that sank Silicon Valley Bank.

The classic shorthand is the repricing gap: assets repricing in a time bucket minus liabilities repricing in that bucket. A positive short-term gap (more assets repricing) = asset-sensitive.

The single most important input is the deposit beta — the fraction of a market-rate change a bank passes through to depositors. A bank with a 30% beta raises deposit rates only 30 bps when the policy rate rises 100 bps. Low betas turn cheap, sticky retail deposits into a powerful margin engine. Critically, deposit beta is not constant: Federal Reserve/Dallas Fed research (Working Paper 2315, 2023) finds betas rise nonlinearly as rates climb, so deposit duration shortens and the funding advantage decays late in a hiking cycle.

How it's used in practice

Drivers of the profile. Asset sensitivity comes from floating-rate or short-dated loans (C&I, credit cards, adjustable mortgages) and short securities books. Liability sensitivity comes from large fixed-rate loan/securities portfolios funded by rate-sensitive deposits or wholesale borrowings. The same logic explains divergent bank behavior: Oldfield Partners contrasts Lloyds (asset-sensitive, with "structural hedges" that smooth NII) against Ally Financial (liability-sensitive, with ~$135bn of multi-year fixed-rate auto loans whose yields lag rising funding costs).

Equity-market application. Bank stocks have historically traded with the 10-year Treasury yield and the slope of the yield curve. Because banks borrow short and lend long, a steepening curve is the textbook tailwind for NIM (Reuters/Investing.com explainer). During the 2013 and 2021 steepening episodes, lenders most reliant on net interest income — typically regional banks — outperformed. This is the practical reason rate-sensitivity profiling is a sector playbook: it tells you which banks to favor for a given rate scenario (asset-sensitive lenders into a hiking/steepening regime; deposit-franchise quality into a cutting regime).

Mitigation tools. Banks use interest-rate swaps and "structural hedges" (laddered fixed-rate receiver portfolios) to dampen sensitivity. Lloyds reported ~£4.2bn of income from its sterling structural hedge in 2024 (up from £3.4bn in 2023) against group net interest income of ~£12.3bn — i.e. on the order of one-third, a derived ratio rather than a figure Lloyds states directly — illustrating how hedging can dominate reported margins.

Adoption, debate & evidence

Rate-sensitivity analysis is universal and regulated — every U.S. bank reports IRR exposure to examiners, and "Sensitivity to Market Risk" is the S in the CAMELS supervisory rating. It is not a fringe technique.

The honest caveats are about the strength of the relationships, not their existence:

  • The bank-stock / yield-curve link is real but noisy. The "10-year up = bank stocks up" heuristic (popularized by outlets like the Motley Fool) holds on average but fails routinely — when rate rises come with credit fears or deposit flight, banks fall with yields rising. Fed research (FEDS Notes, 2025) models bank returns with multiple factors precisely because a single rate factor is insufficient.
  • Asset sensitivity can be a trap. Margin expansion from a low deposit beta is self-limiting: betas climb as the cycle matures. The St. Louis Fed (2024) documented that cumulative deposit betas kept rising even after the Fed stopped hiking in 2023 — loan yields plateaued while funding costs ground higher, compressing NIM into 2024.
  • EVE vs NII can disagree violently. A bank can look asset-sensitive on a 12-month NII basis yet be catastrophically exposed on EVE. At year-end 2022 SVB carried roughly $91bn of held-to-maturity securities at amortized cost (per its 2022 10-K) whose fair value was only ~$76bn — an unrealized loss of ~$15bn, an amount commonly cited as roughly 90% of its total equity (~$16bn), after it had pared back its rate hedges in 2022 (Fed Material Loss Review; JPMorgan). Reported NII looked fine until depositors ran. This is the field's defining lesson: earnings sensitivity and value sensitivity are different risks.

Strengths & limitations

When it works: Rate-sensitivity profiling reliably explains relative bank performance across a rate cycle and flags balance sheets that will be squeezed. Pairing a low-beta deposit franchise with asset sensitivity is a durable competitive advantage.

When it fails: (1) Static gap analysis assumes deposits behave predictably — they don't under stress, when low-beta deposits suddenly flee (the convexity SVB ignored). (2) NII-only views hide duration/EVE risk. (3) The equity heuristic ignores credit risk, which can swamp rate effects in a downturn.

The #1 misuse: treating a bank as asset-sensitive and assuming the benefit is permanent, ignoring rising deposit betas and the unrealized losses building in the securities book. The same rate rise that flatters this quarter's NIM can be hollowing out economic equity.

Sources

Dispute flagged: the "10-year yield up → bank stocks up" relationship is widely cited but empirically noisy and conditional on credit conditions; treat as a tendency, not a rule.