Bank Stocks Are Lagging the S&P 500 by the Most Since 1990

The Invesco KBW Bank ETF (KBWB) is down 14% from its August record and 8% over the past month, while the KBW Nasdaq Bank Index has fallen 13% from mid-August—financials' worst monthly relative performance versus the S&P 500 since 1990. The weakness appears driven primarily by investor rotation and fund flows despite solid earnings, including JPMorgan's 13% adjusted EPS growth and Bank of America's 28% net-income increase. Risks include a flatter Treasury curve, Citigroup's additional $597M loan-loss reserve, declining global M&A volume to $1.23T from $1.8T quarter-over-quarter, and potential AI-enabled migration of low-cost deposits into higher-yielding accounts.

Analysis

The relative drawdown creates a tactical mean-reversion setup only if third-quarter results demonstrate that forward net-interest-income expectations have not rolled over materially. BAC is the highest-quality liquid expression of a steeper-curve rebound given its asset sensitivity and deposit franchise, while JPM's earnings diversity makes it the lower-beta vehicle; C offers greater upside but requires evidence that reserve-building and restructuring costs are peaking. A broad bank ETF trade is less compelling than selective large-cap exposure because regional-bank funding and commercial-real-estate risks can remain a valuation overhang even if money-center earnings hold.

The key near-term mechanism is not reported earnings but forward estimates: a further decline in long-end yields without a commensurate decline in deposit costs would pressure 2026 NII expectations and keep bank multiples capped. Conversely, even modest curve re-steepening can produce disproportionate multiple expansion because positioning is now light and the sector is valued as though margins are already in sustained decline. Over the next 1-3 months, management commentary on deposit betas, loan growth, reserve releases/builds, and investment-banking backlog matters more than backward-looking EPS beats.

The AI-deposit-disintermediation narrative is structurally relevant but likely premature as an earnings catalyst. Consumer inertia, relationship bundling, FDIC preference, and friction around automated cash transfers have historically limited deposit migration; the more immediate threat is that banks preemptively raise rates to defend balances, rather than customers suddenly moving en masse. The contrarian opportunity is therefore a reversal in pessimistic margin assumptions, while the 6-18 month risk is that digital cash-management adoption turns low-cost deposits from a durable moat into a competitively priced product.

Key Decisions for Investors

• Initiate a 3-month long BAC / short KRE pair, sized market-neutral: BAC should benefit more from a benign steepening and has less commercial-real-estate and uninsured-deposit tail risk than regional banks. Target 8-12% relative return; exit if BAC guides to declining NII beyond the next two quarters or if net charge-offs accelerate materially.
• Maintain JPM as the core large-cap bank exposure rather than adding broad KBWB: use a 5-7% pullback into earnings to add, with a 6-12 month horizon. The trade is lower upside than C but offers better protection if deal fees remain weak or credit normalizes faster; reduce on evidence of sustained deposit-cost acceleration without offsetting asset-yield repricing.
• Treat C as an event-driven satellite long only after its next earnings release confirms reserve formation is not worsening and the efficiency program remains on track. A 3-6 month position can offer higher operating-leverage upside than BAC/JPM, but cap sizing given international credit, regulatory, and execution risk; a renewed reserve build would falsify the thesis.
• Do not buy KBWB solely on relative underperformance. Set alerts for a sustained 10-year/2-year steepening of at least 25bp, stable deposit costs, and no broad upward revision to 2026 credit-loss assumptions; absent those conditions, the apparent valuation discount can persist despite strong reported profits.
• For hedging, retain a modest long NDAQ against bank exposure if capital-markets activity recovers: exchange volumes and listing activity can monetize volatility and issuance without the same deposit-beta exposure. This hedge fails if equity-market volumes and IPO issuance remain subdued, so reassess after the next quarterly trading-volume update.