Two things happened to Wells Fargo in mid-September, and the market ended up weighting them backwards. On the fifteenth, at an investor conference, the bank's chief financial officer said that with two months of the quarter behind him, third-quarter net interest margin was tracking better than the company had framed in July — likely down one basis point, possibly flat, against the three-to-four-basis-point decline management had guided. Strip out the markets business, he added, and the banking book's margin would be roughly stable sequentially, with that stability continuing into the fourth quarter . The shares rose on it.
Then the Federal Reserve raised its target range by a quarter point on September 16, its first increase in three years, and the banks were sold as a single instrument. The spread between two- and ten-year Treasuries compressed from roughly 40 basis points at the start of the month to about 25, the financial sector ETF fell around 4% in the days after the decision while the S&P 500 rose 1.6%, and Wells dropped 8.4% from the hike through the following Thursday's close . The stock trades at about $82.97 intraday on September 25, some 15% below its 52-week high of $97.76 .
That sequence is the argument in miniature. A flattening curve is a reasonable thing to sell a generic lender on. It is a poor lens for this one, because Wells' margin has not been a spread story for a year and a half. It has been a mix story — a deliberate, post-asset-cap redeployment of the balance sheet into low-spread markets financing that dilutes the ratio while adding dollars, fees and wallet share. That drag is finishing rather than compounding, the fee engine it bought is compounding instead, and the price is giving credit for neither.
The fourth-largest bank in America, and what it actually sells
Wells is a $2.1 trillion balance sheet and, on assets, the fourth-largest bank holding company in the United States at the end of 2025 . It reports through four segments. Consumer Banking and Lending — branches, checking, cards, mortgage, auto, personal loans — was $37.4 billion of FY2025 revenue against an $83.7 billion total; Corporate and Investment Banking was $19.2 billion, Wealth and Investment Management $16.3 billion, Commercial Banking $12.0 billion, with a $747 million Corporate line and nearly $2 billion of negative reconciling items closing the gap . The consumer bank is the ballast. The other three are where the last two years of change happened.
What changed was regulatory. The Federal Reserve removed the asset cap it imposed in 2018 in June 2025, though other provisions of that consent order remain in place . For seven years Wells could not grow its balance sheet while competitors could. Since the cap came off it has: average loans of $1,026.5 billion in the second quarter of 2026 were up 12% year over year and average deposits of $1,465.6 billion up 10% . That catch-up is also the source of the margin optics that made the stock easy to sell in a rate-driven tape.
The margin compression was bought, not suffered
Reported net interest margin on a taxable-equivalent basis was 2.43% in the second quarter of 2026, down from 2.68% a year earlier — yet net interest income rose 5% to $12.3 billion and total revenue rose 9% to $22.6 billion . A ratio falling while its numerator grows is the signature of denominator growth, and the denominator here was chosen. Management attributes the compression to two things: success in growing interest-bearing commercial and institutional deposits, and expansion of the markets business, where financing balances carry thin spreads but good returns on allocated capital and pull flow business behind them. Average trading-related assets rose 41% year over year, driven primarily by that financing activity .
The shape of the decline supports the mechanism. Margin fell 13 basis points sequentially in the first quarter of 2026, then four in the second, and the mid-September update points to roughly flat in the third . Pure spread erosion in a rising-rate quarter would not taper that way; a low-spread asset class that stops taking incremental share of earning assets produces exactly that curve, because the mix effect on the ratio decays to zero while the assets stay on the books. The full-year frame has not moved either: net interest income of roughly $50 billion for 2026, about $2 billion of it in markets, and noninterest expense of approximately $55.7 billion, both reiterated in mid-September .
The asymmetry the selloff ignores is that Wells' rate sensitivity runs through the short end — commercial and industrial loans and home-equity lines that reprice quickly against a retail deposit base that does not. An instantaneous 100-basis-point parallel move is worth roughly $1.3 billion of net interest revenue and an estimated 4.7% of 2026 core earnings, against under 1% at Capital One, Goldman Sachs or BNY . Wells took its own prime rate to 7.00% effective September 17 ; the deposit repricing that offsets it lags by quarters. Two-to-ten-year slope is the wrong instrument for measuring that, and slope is what moved.
The investment bank is now large enough to change the answer
Corporate and Investment Banking generated $5.4 billion of revenue in the second quarter, up 16%, with Markets up 24% to $2.2 billion — equities alone up 64% — and segment investment banking fees up 36% to $628 million . Firmwide investment banking fees exceeded $900 million, a record for the company, and the share data behind it is specific: leveraged finance at 7.2% year-to-date and a number-three rank, equity capital markets share up 74 basis points to 3.8%, and a climb in U.S. announced-M&A rank from ninth to fourth . By September, management was putting overall investment banking share at 4.3%, against the twos a few years ago, on the back of roughly 150 senior managing directors hired over three or four years .
Wealth and Investment Management grew revenue 13% to $3.9 billion on client assets of $2.41 trillion, up 15% . Commercial Banking grew 6% to $3.1 billion as two years of relationship-manager hiring converted into balances, though it is the segment with the least differentiated franchise and the most direct exposure to middle-market borrowers if tightening bites . Every segment grew. The efficiency ratio came in at 60%, four points better than a year earlier , with headcount at 197,000 — down 15,000 in a year and lower for a twenty-fourth consecutive quarter .
Ballast, a compounding auto book, and a mortgage business still shrinking
The consumer bank did its job without drama: revenue up 6% to $10.3 billion, with credit card up 2% and auto up 33% on originations that rose 41% year over year, helped by becoming the preferred financing provider for Volkswagen and Audi in the U.S. . The unglamorous line is Home Lending, down 7% to $762 million on lower balances and lower servicing income — a business Wells spent years deliberately shrinking, and one that gets no relief from a ten-year Treasury above 5%. It costs roughly half a point of consumer-segment growth now, and I see no near-term mechanism that turns it, but a reader should not read the segment's aggregate 6% as uniform health.
The subtler exposure sits in wealth. Advisory fees are struck on the market value of assets at the beginning of each quarter , so the fourth quarter's fee revenue is already largely set by where markets stood on October 1 — pleasant now, and mechanically punishing after any drawdown, with a full quarter's lag before it shows. That segment runs an 81% efficiency ratio and carries the consolidated advisory-account cash sweep class actions; the high end of reasonably possible losses above accrual across all legal matters was approximately $1.5 billion at June 30 . It is Wells' most equity-market-levered earnings stream, and the one least helped by anything the Fed does.
The loan-loss reserve, not the curve, is the real soft spot
Here I part company with the comfortable version of this story. Realized credit is excellent: net charge-offs of 0.34% of average loans annualized in the second quarter, down ten basis points year over year, and nonperforming assets down $824 million sequentially to 0.77% of loans . Commercial real estate is healing rather than breaking — loans classified as criticized, the examiner grade for weakening credits, fell to $11.8 billion on the mortgage book from $13.4 billion at year-end, with construction down to $1.2 billion from $1.7 billion .
The reserve is another matter. Allowance coverage stands at 1.40% of loans against 1.58% a year earlier , on a book that grew 12%. And the macro scenario blend behind that reserve got materially more optimistic in a single quarter: the forecast fourth-quarter-2026 unemployment rate fell to 4.4% from the 5.0% assumed three months earlier, and the commercial real estate price path for the same quarter flipped to positive 2.7% from negative 7.2% . The same disclosure still carries a negative 7.7% CRE price path for late 2027 . Realized credit genuinely improved, so the direction of the revision is explicable — but it reduced today's reserve, and reversing it is knowable arithmetic. Restoring year-ago coverage on the current book is 18 basis points on $1,026.5 billion of average loans, about $1.85 billion pre-tax, or roughly $0.50 a share after tax on the current diluted base . It would arrive as several quarters of elevated provision against a pre-provision base that is growing faster, not as one charge.
That is the honest downside, and it is bounded. The genuinely dangerous case is not the rebuild alone but its coincidence with a stalled margin — the Fed tightening into a $100-plus oil price until something breaks, Wells surrendering its asset sensitivity on the way back down, and credit costs rising into it. In that world the 17.7% return on tangible common equity printed in the second quarter — flattered by $847 million of venture-capital equity gains and $132 million of discrete tax benefits — compresses toward 15% and the stock revisits the low $70s of its 52-week range . I accept that risk because it requires a recession call, not a curve print, and the credit data does not yet support one.
Ten times earnings for a bank earning seventeen on tangible equity
At $82.97, Wells trades at 10.4 times the $7.99 consensus estimate for FY2027 and at 1.80 times the $46.13 of tangible book per share implied by $139.7 billion of tangible common equity and 3,028.5 million shares outstanding at June 30 . JPMorgan trades at 13.6 times its FY2027 consensus of $25.14, Bank of America at 10.8 times $5.26 . Wells sits roughly where Bank of America does and about a quarter below JPMorgan, on a franchise that earned 16.1% on tangible common equity in the first half — 14.5% in a soft first quarter, 17.7% in the second — against management's stated goal of a sustainable 17% to 18% .
Capital is not the constraint. Common equity tier 1 stood at 10.26% under the binding Standardized Approach against a regulatory minimum and buffers of 8.50%, with total loss-absorbing capacity at 22.81% versus a 21.50% requirement and a 119% liquidity coverage ratio . Wells returned over $9.8 billion in the first half including $7 billion of repurchase , had roughly $22.7 billion left of a $40 billion authorization at mid-year , and raised the quarterly dividend to $0.50 from $0.45 — a 2.4% yield on the current price . Management runs to a 10.0%–10.5% CET1 target and estimates finalized Basel rules would cut risk-weighted assets by roughly 7%, but has explicitly declined to bake that into the repurchase pace until the rule is final . Growth in risk-weighted assets, not authorization, is the live governor on buyback from here.
My own forecast carries FY2027 earnings of $8.61 against the Street's $7.99 ; haircut it by the $0.50 the reserve plausibly gives back and it still sits above consensus. On 1.90 times tangible book — a multiple bracketed by a 16.5% through-cycle return on a roughly 10.25% cost of equity, with and without credit for book growth — applied to the roughly $50.45 of tangible book per share that the current retention and repurchase pace produces a year out, the twelve-month value is $95.83, about 15% above spot and just ahead of the $94.00 consensus target .
Owning the mix shift rather than the curve
The bull case on Wells is not that rates rise, and the bear case is not that the curve flattens. Both are the sector trade, and the sector trade is what set this price. What is actually happening is narrower and more durable: a bank that spent seven years unable to grow has spent fifteen months deploying a balance sheet into markets financing, commercial relationships and an investment bank taking measurable share, paid for it with a ratio rather than with dollars, and is now at the point where the ratio stops falling. The market had that information for one session before a macro print buried it.
The risk worth underwriting is the reserve, and it is about fifty cents of earnings against an estimate that already sits below my own. At 10.4 times FY2027 consensus — a discount to JPMorgan that the return trajectory no longer justifies — the market is charging Wells Fargo for a margin problem it is in the process of finishing. The stock is worth owning into that.
Disclaimer: Tenzing Analytics, Inc. is not a registered investment adviser, broker-dealer, or financial institution, and all content provided by Tenzing Memo is AI-generated for informational purposes only. It does not constitute investment advice, a recommendation, or an offer or solicitation to buy, sell, or hold any security or financial instrument. Content is general in nature, not tailored to any individual's financial situation or objectives, and may contain errors, omissions, or inaccuracies; it should not be relied upon as a primary basis for any investment decision. Any forward-looking statements are subject to risks and uncertainties and may not reflect actual future results. Users should conduct their own research and consult qualified financial, legal, or tax advisors before making investment decisions, and by accessing this content, you acknowledge and accept these limitations.