US Bank Charter Applications: Tracks, Statuses and Grounds for Denial

Since December 2025 the queue for a US bank charter has collected seven fintech applicants. PayPal, Affirm, Klarna, Nissan and Rakuten are on the state industrial loan company track; Nubank and Revolut are on the federal one. As of August 2026 none of them takes deposits on its own balance sheet: most applications are pending, and Nubank holds a conditional approval that carries no right to serve customers.

This page answers two questions of admission rather than any question about products. How the doors differ, and on what grounds a regulator turns an applicant away: the Wise and bunq denials of 2026 are the only public record of what is actually examined.

Concept

Announcing an application is not a licence, and a conditional approval is not a bank. Between filing and the first insured deposit sits a four-rung ladder: application → conditional approval → satisfaction of conditions (capital, deposit insurance and, on a national charter, Federal Reserve approval) → opening. A conditional approval is permission to finish building a bank within a set window, not the bank itself; until it opens, the group's live products run on other institutions' balance sheets and other institutions' licences.

There are three doors, and the owner rather than the product line decides which one.

The state ILC is a Utah or Nevada charter with FDIC insurance under which the parent group stays outside the Bank Holding Company Act by the exemption in the Competitive Equality Banking Act of 1987, so Federal Reserve holding-company supervision does not reach it. That is exactly why commercial and industrial groups file there; the regime is set out in the piece on industrial banks.

The federal de novo national bank from the OCC delivers the full set of powers but pulls the parent under the BHC Act and consolidated Federal Reserve supervision. For a non-financial group that path is closed by construction; for a financial one it is open, and it is the track Nubank and Revolut took.

The national trust charter from the OCC covers custody, fiduciary work and settlement with no right to take deposits or lend, and therefore no FDIC cover. It is the adjacent door Wise filed at; the parameters sit in a separate analysis.

Timelines differ by an order of magnitude. At its current cadence the OCC issues many decisions within 120 days of a complete filing, and from 10 August 2026 the FDIC runs a two-phase process — contingent authorization within 120 days of filing, then full approval over the following 12 months. The typical horizon for reviewing a new charter nevertheless remains 12 to 18 months, and on the state track a decision has historically taken about a year plus another 12 months to open. The wider map of American regimes is in the US licence map.

The applicant grid

Applications are worth comparing on one set of axes: the track determines who supervises and what happens to the parent; the filing date fixes the place in the queue; the status says what the applicant may do today; and the open conditions show where the risk actually sits.

The grid shows an asymmetry between the tracks. The state door has taken five applications and, over eight months, produced no decision on any fintech applicant; the federal door produced three in the same period — one conditional approval and two denials. The state track is slower but quieter; the federal one is faster and harsher, and it is the one that left a public record of the criteria.

The price of entry: how to size the capital

Pending applications carry no public capital figure — the number is set in the decision. The benchmark comes from the same year's approvals of neighbouring captive ILCs: Ford Credit Bank and GM Financial Bank were approved on 22.01.2026 with starting capital of $1.5bn and $667m and tier 1 leverage of 15% or more; Edward Jones Bank on 27.02.2026 with capital of at least $330m, leverage from 9% and a launch planned for 2027; Stellantis Bank USA on 14.05.2026 with leverage from 15% and capital of at least $150m.

The 9–15% spread is not arbitrary: the FDIC calibrates capital to the risk of the model. A captive assembling a retail deposit base from scratch to fund a loan book gets 15%; a brokerage funding secured lending with sweep deposits from its own clients gets 9%. A fintech applicant planning to fund an unsecured consumer book with new retail deposits sits closer to the first group than the second.

The common package of conditions is identical across the wave and raises the entry price more than the headline number does: open within 12 months or the approval lapses; capital and liquidity maintenance agreements from the parent under Part 354 of the FDIC rules; an independent audit; approval of management changes in the early years; dividends requiring FDIC consent for three years. For scale, the adjacent OCC trust charter runs at $6–45m of capital — two orders of magnitude cheaper, but with no deposits and no lending.

What gets an applicant turned away

The 2026 denials are the most valuable part of this map, because approval decisions are written in general terms while denial decisions name defects.

On 21 July 2026 the OCC rejected Wise's application for Wise National Trust on four grounds, none of which touches the business model: a July 2025 Multistate Consent Order for BSA/AML failings; organisers with insufficient knowledge of banking law; a board and management short on fiduciary competence; and no experience operating under 12 CFR 9. The regulator screened the applicant on people and compliance history. The denial does not bar a fresh filing, and the company is preparing a second run.

In August 2026 bunq collected its second consecutive denial, this time on a full de novo. The reasoning was different and no less instructive: raw capital documentation, and the transfer of European credit-loss models onto an American card portfolio. A foreign licence and a European loss record do not by themselves evidence US readiness.

The two denials combine into a single criterion: US readiness is demonstrated before filing, not after. A local team with banking rather than fintech experience; supervisory actions closed rather than open; capital documentation built to American requirements; loss statistics computed on an American portfolio. None of that is about the product.

Reading an applicant's status

Status is read on the same ladder that applies to trust charters: application → conditional approval → satisfaction of conditions → operating bank. The practical consequences for a customer are direct.

Until the bank opens, the group's products are provided by partner banks under existing agreements, and FDIC protection is determined by the bank where the deposit actually sits — often on a pass-through basis with all the caveats of the partner model. A conditional approval does not move existing products onto the new balance sheet automatically and creates no obligation to migrate customers. Rates, limits, documentation requirements and the treatment of non-residents have not been published by any applicant in this wave. Check a specific name against the OCC Weekly Bulletin and Corporate Applications Search, and on the state track against FDIC press releases: the 2026 summary lists contradict one another. Operating banks by jurisdiction are collected separately, and the American perimeter as a whole sits in the US hub.