Ten neobanks built for people who moved — and not one holds a banking licence

September 8, 2026 · 6 min read · ← all posts · the dataset

Two of the last three neobanks added to this directory were built for immigrants. That is not a coincidence, and it is not a small category: money that crosses a border for a person rather than a corporation is one of the largest, least-served flows in retail finance, and the apps chasing it are multiplying.

We track ten neobanks whose stated audience is immigrants and migrants. Seven of the ten were founded this decade. Three arrived in the last two years. And when you line them up against the one question this site exists to ask — who actually holds your money — every single one gives the same uncomfortable answer.

The ten

Zero of ten

Read the right-hand column again. Six run on a partner bank's charter. Three hold e-money or payment-institution permissions. One is licensed as a money-services business and VASP. None of the ten holds a banking licence of its own — against 132 of the 381 neobanks in the dataset overall, which is roughly a third.

This is not an accusation. Renting a licence is how you launch a specialised product in three markets without spending five years and a great deal of capital becoming a bank, and several of these companies are transparent about the arrangement. It is also, structurally, the arrangement that fails most often. Of the six neobanks that left this directory this year, three died because the institution underneath them withdrew — not because customers left.

Which produces the thing worth sitting with: the customers with the least margin for error are concentrated on the structures with the most intermediaries between them and their money. A frozen account is an inconvenience to someone with a second bank and a credit card. To someone sending a month's wages home, or newly arrived without a local credit history, it is the whole month.

Why the model keeps winning anyway

Because incumbents genuinely do not serve these customers. A new arrival with no domestic credit file is invisible to underwriting built on one. Remittance pricing has been quietly punitive for decades. Documentation requirements assume a life that has been lived in one country. Every one of the ten exists in a gap a licensed bank left open, and their pitches are specific in a way generalist banks are not: Zolve underwrites arrivals in the United States on their home-country history, Majority bundles calling and community services with the account, Félix moves money through a messaging app, Yolat runs the Canada–Nigeria corridor on rails licensed at both ends.

Two of the ten — Félix and GetPlu — settle on stablecoins rather than correspondent banking. Expect that share to rise, since it attacks precisely the cost the incumbents never fixed. It also swaps one set of intermediaries for another, which is a trade worth understanding rather than cheering.

The honest limits of this cut

Ten is a small sample, and the boundary is fuzzy: dozens of remittance-adjacent apps and multi-currency accounts serve migrants without describing themselves that way, and we tag the audience a company states rather than inferring one. Reported user figures barely exist here — most of these companies publish nothing, and we do not estimate. What the cut does show reliably is structural: a young, fast-growing cohort, uniformly built on rented rails, serving people for whom a rail failure is not an inconvenience.

If we are missing one — and in this category we probably are — tell us. Both of the most recent additions came from readers.