Forbright Calls It 'Terrific' — But the Real Bet Is That Branch Deposits Are a Melting Iceberg

Forbright, Inc. is a $936 million market-cap bank that spent the third quarter of 2026 doing something unusual: growing its loan book 9.5% in three months, raising a billion dollars of deposits in the same three months, and then telling investors it didn't need to raise deposit rates after the Fed hiked in September. Management's own framing was self-aware. “I'm somewhat superstitious by nature, so I'm always a little reluctant to call a quarter terrific. But this time, the facts certainly merit that description to my mind.” — John Delaney, Chief Executive Officer (CEO) · 2026-10-29

The Quarter That Earned the Adjective

The arithmetic is genuinely strong. Net income was $20.6 million, or $0.41 per diluted share, up $16.5 million and more than 400% from the second quarter. Net interest income rose to $72 million from $63 million, and pre-provision net revenue climbed nearly 60% sequentially. Loan growth was 24% year-over-year and 9.5% quarter-over-quarter, led by health care, though the CEO was careful to note that health care probably won't be the largest contributor next quarter — the point being that all six national lending verticals are producing, not one.

Credit is the number that matters most for a bank of this shape. Net charge-offs were just $1.2 million, down from $2.7 million, and the core net charge-off rate across the six national lending strategies was 3 basis points, versus 8 basis points last quarter. Management reiterated a 25-basis-point annualized through-the-cycle forecast and conceded it is beating it. The efficiency ratio fell to 69.6%, roughly 8 percentage points better than the prior quarter.

One honest caveat got buried in the prepared remarks: NIM was flat, not wider, and roughly 10 to 11 basis points of that flatness was self-inflicted — excess liquidity held ahead of a branch sale inflating the denominator of the loan-to-deposit math.

The Knobs: Pricing Power in a Deposit War

Here is where Forbright diverges from the crowd. Across this reporting cycle the shared banking theme has been the cost of money — Fed rate hike chatter and rising funding costs thread through regional bank calls from LARK to RNST, where core deposits and total deposit costs dominated the Q&A. Forbright went the other way. “We held our gross savings rate steady even as the Fed raised rates in September.” — Donald F. Cole, Chief Financial Officer (CFO) or Senior Finance Executive · 2026-10-29 Roughly 90% of loans and 85% of deposits are floating, so the bank is structurally asset-sensitive; holding the base rate while assets reprice is the mechanism by which management believes NIM widens in the fourth quarter. “As we sit here today, we would expect our NIM in the fourth quarter to, in fact, be wider.” — John Delaney, Chief Executive Officer (CEO) · 2026-10-29

The engine behind that is the digital deposit platform and its new promotion engine, which pulled in over $1 billion of deposits — up 24% quarter-over-quarter — at an acquisition cost management calls attractive. That scale gave them optionality on deposit pricing, and the CEO is now openly flirting with the other dial: “So I think we're in that kind of position where we have a little bit of pricing power.” — John Delaney, Chief Executive Officer (CEO) · 2026-10-29 The plan is to test widen spreads on next quarter's origination vintages. He was careful — repeatedly — to call any widening "very modest," five or ten basis points, and to acknowledge a "very tight spread environment." That caution is the tell that this is an aspiration, not a booked margin.

Also worth noting: the deposit ratio rose 4 percentage points to 87.3% against a low-90s target. That is the less glamorous, more reliable path to a better NIM — simply lending out the funding they already have.

An AI Thesis Wearing a Bank's Clothes

The genuinely unusual part of this call was not the quarter. It was the CEO's argument that the deposit business is on the front end of a technology adoption curve. Legacy branch deposits, he said, “resemble a slow melting iceberg.” — John Delaney, Chief Executive Officer (CEO) · 2026-10-29 The supporting math is striking: there is roughly $10 trillion of U.S. consumer savings priced near 40 basis points that, in his view, should be earning 350 to 400.

“There's about $10 trillion of consumer savings accounts in the United States that are priced at 40 bps, and those people should be getting 350 to 400 bps.” — John Delaney, Chief Executive Officer (CEO) · 2026-10-29

What makes this more than a slogan is that Forbright is shipping product against it. The company launched a virtual card for AI agents — a separate card from the core debit card, with spend limits, merchant-category restrictions, single-transaction caps, and a human-in-the-loop second-factor approval inside the digital banking ecosystem. It is the same idea Mastercard is describing as Agent Connect and that WEX is selling as a virtual card offering — the difference is that Forbright is aiming it at consumer deposit-gathering rather than B2B payments. The broader Agentic AI wave is clearly now a banking topic, not just a software one.

The risk, which management raised itself, is security concerns. The CEO's answer was pointed: those concerns are "ultimately manageable," and much of the industry handwringing is incumbents protecting a cheap funding base. He even floated a policy idea — regulate agents as fiduciaries, accomplishing by technology what the CFPB was built to do by rule. Provocative, and entirely unquantified in any forecast.

If something cannot go on forever, it will stop. And I think what can't go on forever is consumers leaving tremendous value on the table because moving their money was inconvenient and hard in the past, and technology will eliminate that inconvenience.

What Actually Changed

Two quieter items are the real structural news. First, the branch sale: Forbright is selling its legacy Washington, D.C.-area branches and associated deposits to Trustar Bank, with regulatory approval expected in the fourth quarter and closing in the first quarter of 2027. That is the completion of a pivot — the company will be a digital-only deposit gatherer. It costs near-term NIM (higher-cost digital funding replaces branch deposits) and saves opex that management intends to redeploy. Second, the tax line, where deferred credit mechanics matter more than they should: the balance fell to $36 million from $49 million after a purchase accounting adjustment reversed $2.7 million of benefit. Guidance is a 20% rate in Q4 and 19% for 2027, each net of roughly 9 points of deferred credit amortization — a mechanical earnings tailwind that will flatter 2027 EPS growth.

What faded is equally telling. Solar servicing — an $8 billion master-serviced book — was reframed from growth story to pass-through noise and rate hedge: higher long rates slow HUD gain-on-sale originations but stretch the life of the serviced portfolio. And the legacy Congressional Bank commercial real estate book was openly named as the source of the bank's only noisy credit. Neither is a retreat so much as a demotion.

So: a genuinely excellent quarter, a credible near-term NIM catalyst, and a genuinely contrarian long thesis that will not show up in earnings for years. The question investors must underwrite is whether Forbright's deposit-cost advantage survives a real hiking cycle — the CFO conceded the next hike will carry "a higher beta than zero" — and whether eliminating the branch network saves more in expense than it costs in blended funding. That is the whole bet, and it is a bet on 2027, not on a terrific third quarter.