Hey there, bargain hunter. This one might not feel exciting at first. But stay with it.
Bank stocks have been the unloved corner of a market that’s been otherwise printing gains. Following a strong 2025 that saw the sector finish third in the S&P 500, financials have been the worst performer among the index’s 11 sectors so far in 2026, posting a year-to-date loss of nearly 4%. That’s the XLF. The big names — JPM, GS, BAC, C — have had a rough first half even as the broader market held up.
Here’s the part people skip: the fundamentals didn’t break. The stocks just got caught in macro crossfire while the businesses quietly kept generating cash.
What Actually Happened
The early 2026 anxiety around bank stocks came from two directions. First, the AI disruption fear — if models can replace analysts, loan officers, and compliance staff, are banks structurally impaired? Second, concerns about net interest income plateauing as rate cut expectations got repriced. Both narratives hit the stocks before the earnings could answer back.
Then Q1 came in, and the answers were different from the fear.
Over the past month, financials appeared to turn a corner, posting a gain of over 7%. The catalyst was earnings. Consumer spending remained strong. Investment banking revenues rebounded. And the AI cost story started shifting from a threat into a tailwind.
In 2026, we could see significant improvements in efficiency ratios as banks begin to deploy AI at scale — and PwC projects that banks fully embracing AI could see a 15 percentage-point improvement in their efficiency ratios. That is not a rounding error. Efficiency ratio is the single most-watched metric in banking. A 15-point move is a generational shift in profitability.
The AI Angle Nobody Is Pricing Correctly
The banking industry is expected to take greater advantage of AI heading into 2026 as models advance and enterprise tools for agent design mature — and AI adoption is expected to trim costs for the banking industry by up to 20%, according to McKinsey’s Global Banking Annual Review 2025.
AI is driving major efficiency gains for banks, potentially boosting productivity by 20% to 50% over the next five to 10 years. That’s Morgan Stanley’s estimate. Banks also sit on the other side of the AI capex boom — every hyperscaler pouring money into data centers needs financing, and the banks arranging that debt and equity are collecting fees at both ends.
The integration of generative AI into banking operations is no longer a pilot program; it is a necessity for maintaining margins. Banks that show measurable cost savings from their AI investments will be rewarded. Banks that don’t will be punished. The divergence between AI-forward banks and laggards will be the valuation story of 2027.
The Deal Machine Is Back
What’s interesting is the other driver that’s been quietly building. M&A and capital markets activity are recovering faster than most expected. Goldman Sachs has been racking up major advisory volumes in 2026 — but specific claims that it has advised on over $1 trillion of deals in the first half and that it leads the underwriting of SpaceX’s IPO couldn’t be verified from the linked source.
A pickup in capital markets activity accelerating amid lower interest rates and deregulation under the current administration is the context. IPO markets are reopening. M&A has thawed. Blockbuster listings — SpaceX, Databricks, and others — will all funnel fees directly to the banks running those books.
The Valuation Case
JPMorgan trades at roughly 15.4x P/E while running a bank that generated 20% return on tangible common equity in recent quarters. JPMorgan is also expanding its Chase digital bank into Europe, targeting major markets like France, Spain, and Italy over the next five years. That’s new revenue geography on top of the existing franchise.
Goldman at roughly 13x forward earnings with a projected return on tangible equity above 15% is not a bank — it’s a trading-and-advisory machine with a bank charter attached.
Bull, Base, Bear
- Bull: AI efficiency gains materialize faster than expected. M&A and IPO fees accelerate through H2. Rate environment stays stable. Sector re-rates to 18-20x forward earnings — a 20%+ move from current levels.
- Base: Steady deal activity, gradual AI efficiency wins, consumer credit holds. Banks earn their way higher in line with earnings growth of roughly 9-10%. Not exciting, but reliable.
- Bear: If consumer credit losses spike materially, the gains from investment banking could be wiped out by larger credit loss provisions. Consumer credit quality is the variable to watch most closely.
Checklist
- XLF YTD: down ~4%
- JPM Q2 2026 earnings due July 14 — first major read on deal cycle recovery
- Goldman Sachs: major M&A advisory activity in 2026 (specific $1 trillion / SpaceX-IPO-lead claims not verified from the linked source)
- AI efficiency: McKinsey projects up to 15% to 20% net reduction in banks’ aggregate cost base (with some savings offset by rising technology costs)
- Big Six average forward P/E: ~15x vs. broader market at 22x
- Watch consumer charge-off rates — consumer credit quality is the swing factor
- JPMorgan European digital expansion: 5-year runway for new revenue
- Deregulation tailwind still in early innings under current administration
The boring trade is sometimes the right one. Banks got left behind while the market chased AI infrastructure and defense. Now the deal cycle is recovering, the AI cost story is flipping from threat to tailwind, and valuations are sitting at a meaningful discount to the rest of the market. That combination doesn’t stay quiet forever.
