Let me concede something upfront: I do not have the primary filing for the Fasset valuation round in the grounding envelope this desk writes inside, so I am not going to invent numbers around it. What I do have is the actual definitional plumbing of the phrase "stablecoin neobank" — a term that keeps showing up in headlines like this one and almost never gets defined at the layer that decides whether it means anything. Anchorage Digital holds an OCC Federal Trust Charter — the first crypto bank in the United States. Coinbase Custody is a NY DFS Trust Company. Those two facts, from the public record, are the entire measuring stick.
What the Numbers Actually Say About the Neobank Label
Here is the receipt. A "neobank" in the fintech sense is a consumer-facing interface that sits on top of a banking rail — usually a partner bank that holds the actual charter. The neobank does the UX, the ledger, the card issuance, the notifications. The partner bank does the thing that requires a regulator. Two entities, one wrapper.
A "stablecoin neobank" claims to do the same shape of thing, except the underlying rail is not a partner bank — it is a stablecoin ledger. USDC, USDT, PYUSD, whatever the specific denomination happens to be that quarter. The pitch is that the user gets a debit card, a phone-app balance, and cross-border settlement, and the plumbing underneath is on-chain instead of ACH-and-SWIFT.
That is the marketing layer. Read carefully.
Now read the custody layer. When a user deposits fiat into a stablecoin neobank and receives an in-app balance denominated in a stablecoin, someone is holding either (a) the customer's fiat before it converts, (b) the customer's stablecoin after it converts, or (c) both, at different points in the flow. That "someone" is a legal entity with a specific regulatory status. In the U.S. context, the possible statuses are narrow and named. Anchorage Digital operates under an OCC Federal Trust Charter — the first crypto bank in the United States, granted in January 2021. Coinbase Custody Trust Company operates under a NY DFS Trust Company charter, granted in 2018. Fidelity Digital Assets operates under a NY DFS Trust as well.
Those three statuses — OCC Federal Trust, NY DFS Trust, and equivalents in a handful of other state regimes — are the only ones in the U.S. that let a legal entity hold customer crypto assets in a way that meaningfully resembles what a chartered bank does with customer dollars. Everything else is either a money transmitter license (MTL) stack, an offshore trust, or a self-declared "custody solution" that legally is neither.
So when a headline says "stablecoin neobank raises at [X] valuation," the question that decides whether the label is real is: which of those charters is under the product? If the answer is "none — funds are held by a partner exchange offshore" then the neobank framing is a UX layer over an unlicensed pool. If the answer is "OCC Federal Trust" or "NY DFS Trust," the framing has real weight. The word "neobank" is doing enormous load-bearing work in a headline where the difference between those two answers is the entire investment thesis.
Nobody writes about which one.
What Nobody Mentions About the Custody Charter Underneath
OK so here is where it gets really interesting — and I love this detail, so let me digress for two paragraphs because it is the actual point.
The reason "custody charter" is the load-bearing definitional layer, and not "banking license" or "money transmitter registration" or "e-money institution," is that stablecoins are legally property, not deposits. When you deposit USD into Chase, Chase owes you USD as a debt liability and the FDIC insures the first $250,000 of that debt. When you deposit USDC into a custodian, the custodian is holding your USDC as bailed property. If the custodian fails, your USDC is (in theory) not part of the bankruptcy estate — it belongs to you and is returned to you, subject to the specific trust law of the custodian's jurisdiction.
That "in theory" is doing the heavy lifting. It is only true if the entity holding the assets is legally structured to segregate customer property from the entity's own balance sheet. A Federal Trust Charter under the OCC requires that structure by statute. A NY DFS Trust Company requires it. A generic money transmitter license — the license under which most consumer crypto apps in the U.S. actually operate — does not, and in the post-Celsius, post-FTX, post-Voyager wreckage of the 2022 cycle it became painfully clear how consequential the difference is.
Now overlay that on the exchange data this desk actually has in the grounding.
Binance publishes a proof-of-reserves attestation dated 2025-03-01, is licensed in Dubai (VARA, full, tier 2), France (AMF, limited), and Italy (OAM, limited). Bybit publishes PoR dated 2025-03-12, licensed in Cyprus (CySEC) and Dubai (VARA). OKX publishes PoR dated 2025-03-01, licensed provisional in Dubai (VARA) and full in Bahamas (SCB). Bitget: PoR 2025-02-20, Lithuania (FCIS) and Poland (KNF). MEXC: PoR 2024-12-10 marked as partial reserve status, offshore Seychelles.
None of those is a U.S. trust charter. None of them lets a consumer-facing "neobank" wrapper claim that customer stablecoins are held under a legal structure equivalent to Anchorage or Coinbase Custody Trust. What they let you claim is that a third-party attester walked through a snapshot of on-chain balances and cross-referenced them against a claim about liabilities on a specific date — with the sophistication of that liabilities check varying enormously between attestations. PoR is a real check. It is not a trust charter. Conflating the two is the specific error that the phrase "stablecoin neobank" almost always launders past readers.
That is the layer nobody mentions in the funding-round coverage. Not the round size. Not the backer's identity. The custody charter under the wrapper.
The Real Cost of Reading Valuation as Trust Signal
Here is where you put a dollar figure on the gap, and it is the part I care about the most.
A retail user reads a headline: "Stablecoin neobank hits [$X billion] valuation, backed by [major financial institution]." The user reads that as an implicit trust signal — the reasoning goes, "a major institution would not back a company that could lose my money, and a company at this valuation must have serious regulatory infrastructure." That reasoning is wrong at both steps, and the cost of it being wrong is not theoretical.
Cost model, plain arithmetic. Suppose the user deposits $10,000 into the neobank, receives a USDC balance in-app, uses the card for spending, and does not withdraw. The user's exposure is now conditional on: (1) the custodian actually holding the USDC in segregated form, (2) the neobank's off-chain ledger correctly reflecting the on-chain balance, (3) the USDC issuer maintaining redemption, and (4) the wrapper's failure mode being handled by an actual trust law that returns the asset rather than a bankruptcy estate that liquidates it against creditors.
If the custodian under the wrapper is Anchorage (OCC Federal Trust) or Coinbase Custody (NY DFS Trust), item (4) resolves cleanly — trust property, returned to beneficiaries, out of estate. Failure cost approximates the time-value of the delay plus any processing frictions. Small.
If the custodian is a foreign exchange with a PoR attestation but no U.S. trust charter, item (4) is jurisdiction-dependent and, in the recent record, often catastrophic. FTX customers who thought their funds were "custodied" learned in November 2022 that the omnibus commingling structure meant they were unsecured creditors. Celsius customers learned the same thing in July 2022. Voyager. BlockFi. Each of those failures produced customer recovery percentages that varied from roughly 30% to roughly 80% of nominal, over recovery windows measured in years. On a $10,000 deposit, the expected-value hit ranges from $2,000 to $7,000 and the liquidity hit is total for 18 to 36 months.
That is the price of misreading valuation as trust signal. A billion-dollar valuation from a strategic backer is not a trust charter. It is a bet on the wrapper's growth. Those are different bets and they resolve on different ledgers.
The exchange data in this desk's grounding underlines the point. Binance runs $18.5B daily volume, 350 supported coins, tier-2 licenses across three jurisdictions, PoR attestation refreshed roughly quarterly — and none of that would matter to a bankruptcy court in New York if the entity holding the pool failed and the receiver had to decide whether customer assets were property or debt. Volume is not custody. License is not trust. PoR is not solvency. Those distinctions cost real money to real people in every prior cycle.
If You Only Remember One Thing
When a headline calls something a "stablecoin neobank," the only question that decides whether the label is real is: what is the custody charter of the entity holding customer stablecoins? OCC Federal Trust and NY DFS Trust Company are the two U.S. statuses that carry actual weight. Everything else is a UX layer over a legal structure that behaves very differently in a failure scenario.
I would reverse this framing if a stablecoin-neobank wrapper published, on a public disclosure page, the custodian entity, the custodian's charter number, the segregation legal opinion, and a proof-of-reserves attestation that reconciles against a proof-of-liabilities check performed by a named auditor within the prior 90 days. Until that disclosure stack exists as the default and not the exception, "stablecoin neobank" is a marketing phrase and the valuation number in the headline is a marketing number. Read the custody charter, not the round size.
FAQ
Is "stablecoin neobank" a defined regulatory category?
No. It is a marketing term, not a regulatory one. There is no jurisdiction — U.S., EU, UAE, Singapore — that issues a "stablecoin neobank" license as such. The underlying legal statuses are always something else: a state-chartered trust company, an OCC federal trust, an EU e-money institution, a Dubai VARA licensee, or a stack of state-level money transmitter licenses. The neobank framing is the consumer interface; the license underneath decides the substance.
What is the difference between a custody charter and a money transmitter license?
A custody charter — OCC Federal Trust for Anchorage Digital, NY DFS Trust for Coinbase Custody and Fidelity Digital Assets — authorizes an entity to hold customer assets as bailed property, legally segregated from the entity's own balance sheet. A money transmitter license authorizes the movement of value from A to B and does not require segregation of held customer assets in the same statutory way. In a failure, the difference decides whether customer funds are returned or become part of a bankruptcy estate.
Does proof-of-reserves prove that a stablecoin neobank is solvent?
No, and this is the most common misreading in the category. Proof-of-reserves attests that a set of on-chain wallet balances exists on a specific date. It does not, by itself, attest to liabilities. Binance's PoR is dated 2025-03-01; Bybit's is 2025-03-12; OKX's is 2025-03-01; Bitget's is 2025-02-20; MEXC's is 2024-12-10 and flagged partial. Any of those figures can be true while the entity is functionally insolvent if liabilities exceed the attested reserves. Reserves without liabilities is a snapshot, not a solvency proof.
If I use a self-custody wallet, do any of these custody-charter questions apply to me?
No — that is the entire point of self-custody. When keys to a Ledger, Trezor, or GridPlus Lattice1 device are held by the user, no third-party custodian is in the failure path. The tradeoff is that the user assumes full operational responsibility: seed backup, firmware verification, transaction signing hygiene. Self-custody removes the custodian-failure risk and replaces it with user-error risk. Which risk is preferable is a function of the user's operational discipline, not of the neobank wrapper's marketing.
Why does the U.S. specifically matter when the headline mentions a UAE-domiciled fintech?
It matters because the U.S. trust charter framework is currently the strictest publicly documented custody standard, so it functions as a comparative benchmark for what "real" charter-backed custody looks like. A UAE-domiciled wrapper operating under Dubai VARA is not obligated to meet the OCC standard, but a reader trying to assess whether the entity's custody status is meaningful can use the U.S. framework as a measuring stick. If the disclosed custody structure would not survive a comparison to Anchorage or Coinbase Custody's statutory obligations, that is a data point.
Which stablecoin does the neobank use — does it affect risk?
Yes, and materially. The stablecoin issuer is a separate counterparty from the neobank and the custodian. USDC is issued by Circle, redemption-backed against a disclosed reserve, attested monthly. USDT is issued by Tether, with a different reserve composition and disclosure cadence. PYUSD is issued by Paxos Trust, a NY DFS trust entity. A user's exposure is the joint probability of the issuer holding par, the custodian holding the token in segregated form, and the neobank's ledger correctly reflecting the balance. Three independent points of failure, not one.
What disclosures should a stablecoin neobank publish for its custody claim to be credible?
At minimum: the legal name and jurisdiction of the custodian, the custodian's charter or license number, a written legal opinion on customer-asset segregation in bankruptcy, and a proof-of-reserves attestation reconciled against a proof-of-liabilities check by a named auditor within the prior 90 days. That stack is not standard practice in the category today. When it becomes standard, the neobank label starts to carry real meaning. Until then, treat the marketing language as decorative and read the underlying legal structure directly.