The Crypto Bank Race: Who Gets to Hold Your Coins in 2026
In 2026 a dozen crypto firms raced to become chartered trust banks, and community lenders sued to stop them. Here is what an OCC charter really guarantees about who holds your coins.
On the weekend before this was written, Bitcoin fell to roughly $55,252, its lowest level since September 2024, before clawing back above $60,000 by the time New York desks opened on Friday. Ether traded near $1,560. A drawdown like that does two things at once. It wipes out paper gains, and it reminds everyone that a coin you do not actually control is a coin you have only been promised. When prices fall and withdrawals spike, the question stops being which token to hold and becomes a plainer one: who is holding it, and under what law.
That question has an unusually concrete answer in 2026, because a dozen crypto firms spent the year racing to become something they could not have been three years ago: a federally chartered bank. Not a bank that takes your paycheck and lends it back out, but a national trust bank, a limited-purpose institution supervised by the Office of the Comptroller of the Currency whose entire reason to exist is holding assets for other people. Anchorage got there first, back in 2021. Then in December 2025 the OCC waved through five more in a single day. By late October 2026 Ripple had a final charter, Galaxy had filed its own application, and the community-banking lobby had sued the regulator to make the whole thing stop.
This piece is a comparison, not a cheerleading exercise. A charter is one row in the custody table, not the whole table. It changes some things about who holds your coins and leaves others exactly where they were. The macro backdrop matters too, because custody questions always get loudest in a sell-off, and this one arrived in the middle of a brutal one that had little to do with crypto plumbing and a lot to do with a jittery December rate decision and a repricing of risk across every market. So it is worth being precise about what the 2026 charter race actually buys, and for whom.
What a National Trust Bank Charter Actually Is
Start with the thing being handed out, because most coverage skips it. A national trust bank is a federally chartered institution that does fiduciary and custody work and almost nothing else. It is supervised by the OCC, the same agency that oversees the largest US commercial banks, but it operates under tight limits. The crypto firms in this cohort are chartered as trust banks that do not take insured deposits, do not make commercial loans, and are not covered by the Federal Deposit Insurance Corporation. The whole point is safekeeping: holding client assets, keeping a register of them, and standing behind them as a fiduciary.
That last word carries the weight. A fiduciary custodian has a legal duty to act in the client’s interest and to keep client property segregated from its own. Compare that to an ordinary crypto exchange, which in most cases holds your coins as a counterparty under its terms of service, not as a fiduciary holding your specific property. The charter does not change the blockchain or the private keys. It changes the body of law that wraps around the custodian, and that wrapper is what actually gets tested when something goes wrong.
The word “bank” does a lot of misleading work here, so it helps to say what these institutions are not. They are not deposit-taking banks; you cannot open a checking account at Ripple National Trust Bank, and there is no insured balance the government stands behind. They are closer to the old-line trust companies that have safeguarded securities and estates for more than a century, updated for an asset that settles on a public blockchain. The novelty is the asset, not the function, which is precisely the argument the Comptroller leans on when he says there is no reason to treat digital assets differently.
There is a federal angle that matters for firms that already hold state trust charters. Before 2025, a custodian like a New York limited-purpose trust company answered to a single state regulator. A national charter swaps dozens of potential state regimes for one federal supervisor, which is why several applicants are existing state trusts converting upward rather than brand-new entities. The OCC frames this as ordinary banking plumbing. “Custody and safekeeping services have been happening electronically for decades,” Comptroller Jonathan Gould said when the December 2025 approvals landed. “There is simply no justification for considering digital assets differently” (Banking Dive).
The 2026 Field: Who Is Racing for a Charter
The cohort is bigger than any single headline. Some firms already hold charters; some are state trusts converting up; some have only just filed. The table below maps the field as of late November 2026, with the standing caveat that applications move weekly and a filing is not an approval.
| Firm | Status (late Nov 2026) | What it becomes with a charter | Note |
|---|---|---|---|
| Anchorage Digital Bank, N.A. | Chartered Jan 2021 | First federally chartered crypto bank | Operates nationwide; the template the others followed |
| BitGo Bank & Trust, N.A. | Full charter Jan 2026 | Public, federally chartered custodian | First publicly traded firm of its kind (NYSE: BTGO) |
| Circle (First National Digital Currency Bank) | Conditional Dec 2025 | Trust bank for USDC reserves plus custody | Stablecoin issuer going prudential |
| Ripple National Trust Bank | Final approval Oct 2026 | Trust bank for RLUSD reserves plus custody | Charter to hold the roughly $2.6B behind RLUSD |
| Paxos Trust Company, N.A. | Conditional Dec 2025 | State trust converting to national | Stablecoin and tokenization infrastructure |
| Fidelity Digital Assets, N.A. | Conditional Dec 2025 | Trust-bank arm of a legacy asset manager | Self-custodies its own spot Bitcoin ETF |
| Crypto.com | Conditional 2026 | Exchange-affiliated trust bank | Trading group seeking a chartered custody arm |
| Galaxy National Bank | Applied Oct 2026 | Custody, trading, staking and advisory | Novogratz firm; no deposits or lending |
| Coinbase and others | Pending or expected | Varies | Largest US ETF custodian; charter path still open |
The spread is the story. You have a stablecoin issuer (Circle) seeking a charter to sit its reserves inside a prudential wrapper, a payments company (Ripple) doing the same for RLUSD, a legacy asset manager (Fidelity) formalizing an arm it already ran, and a trading-heavy shop (Galaxy) explicitly bundling custody with trading and staking under one federal roof. Those are different businesses asking for the same license, which is exactly what the charter’s critics object to. The December 2025 batch alone (Circle, Ripple, Paxos, BitGo and Fidelity Digital Assets) moved the count of chartered or conditionally chartered crypto custodians from a curiosity to a cohort.
How the Gates Opened
The race did not happen because crypto suddenly earned trust. It happened because two supervisory barriers fell in quick succession. The first was SAB 121, a 2022 Securities and Exchange Commission staff bulletin that forced any firm safeguarding client crypto to record it as a liability on its own balance sheet. For a bank, that treatment was close to prohibitive, because it inflated the balance sheet and the capital required against it. In January 2025 the SEC rescinded it with SAB 122, and bank custody of crypto went from uneconomic to merely complicated more or less overnight (ABA Banking Journal).
The second shift was a pair of OCC interpretive letters in spring 2025 confirming that national banks may custody crypto, buy and sell it on a custody client’s behalf, and outsource execution, all without asking for prior supervisory sign-off. Put those together with the GENIUS Act, the stablecoin law signed in July 2025 that gave dollar-token issuers a federal reserve-and-disclosure regime, and the incentives lined up. A stablecoin issuer now had a reason to want a prudential charter, because the law pushed reserves toward cash and short-term Treasuries with monthly attestations; a bank now had permission to hold the coins. The charter race is the predictable result of clearing both hurdles in the same twelve months.
It is worth being clear that none of this required Congress to pass a crypto custody law. The gates opened through an accounting reversal at the SEC and interpretive guidance at the OCC, not a statute that defines how digital-asset custody must work. That distinction matters later, because a rule made by guidance can be unmade by guidance, and a charter granted under contested authority can be challenged in court. The 2026 custody regime rests on administrative footing more than legislative bedrock, which is part of why the community banks saw an opening to sue.
Ripple’s Win and Galaxy’s Bid
Two events in a single late-October week showed how fast this is moving. First, Ripple received final, unconditional OCC approval for Ripple National Trust Bank, upgrading the conditional nod it had won in December 2025. The charter lets Ripple custody the reserves behind RLUSD, its dollar stablecoin, inside a federally supervised trust rather than parking them with a third party (PYMNTS). For a stablecoin, the location of the reserves is the product, so moving them into a chartered bank is less a vanity badge than a structural change in who stands behind the token.
Days later, Galaxy Digital filed its own application for Galaxy National Bank, a proposed New York trust bank that would, if approved, hold custody, trading, staking and fiduciary advisory under one OCC license while taking no deposits and making no loans (Banking Dive). “Digital assets are becoming an increasingly important part of the global financial system,” CEO Mike Novogratz said, calling the charter “an important next step, allowing us to bring more of our services under a single federal framework.” Galaxy already holds more than fifty licenses across the US, Europe and Asia, including a broker-dealer registration and New York’s BitLicense; the charter would fold a pile of them into one supervisor.
The pattern across both is consolidation. Firms that grew up stitching together state money-transmitter licenses, trust charters and offshore entities now see a single federal charter as the cleaner way to operate at scale. That is the bull case for the whole race. It is also, as the next few sections argue, easy to oversell.
The Two Pioneers: Anchorage and BitGo
Two names anchor the field because they got there first and they got there differently. Anchorage Digital Bank became the first federally chartered crypto bank in January 2021, years before the current wave, after the OCC granted it a national trust charter that let it run custody across all fifty states under one federal supervisor (CoinDesk). For most of the time since, it was the lonely proof that the model could exist at all. Its pitch has always been security engineering: hardware security modules, biometric approval quorums, and a careful governance of exactly who can authorize a withdrawal.
BitGo arrived by a different route and hit a different milestone. Founded in 2013 as a multisignature pioneer, it converted its trust company into BitGo Bank and Trust, N.A. in early 2026 and then did something none of the others had: it went public, listing on the New York Stock Exchange as the first publicly traded, federally chartered digital-asset custodian. A public listing means audited financials and quarterly disclosure, a layer of scrutiny a private custodian never faces, which is its own kind of custody assurance even though it says nothing about how the keys are held.
The contrast between the two is a useful miniature of the whole race. Anchorage sells the vault, the cryptographic and operational hardening of how coins are stored. BitGo increasingly sells the institution, a chartered and now publicly accountable entity whose books you can read. A careful custody buyer wants both, and the charter cohort is, in effect, a competition over which combination of vault and institution wins the mandates. Neither, it bears repeating, turns a held coin into an insured deposit.
A Charter Is One Row, Not the Whole Table
Here is where the comparison has to get honest. A national trust bank charter is a strong form of custody, but it sits on a spectrum, and moving up that spectrum trades one kind of risk for another. The table below lines up the main models by two things that usually get collapsed into one: who technically controls the keys, and what legal protection you actually have if the holder fails.
| Custody model | Who holds the keys | Legal protection if the holder fails | Main risk you keep |
|---|---|---|---|
| Self-custody (hardware or software wallet) | You, alone | None needed; there is no third party to fail | You lose the keys, or you get phished |
| Exchange account | The exchange, as counterparty | Terms of service; often a general creditor claim in bankruptcy | Exchange insolvency or hack |
| State-chartered trust custodian | The custodian, as fiduciary | State trust law; client assets segregated | Regulator quality varies by state |
| OCC national trust bank | The custodian, as fiduciary | Federal supervision, segregation, fiduciary duty | Private insurance only; no FDIC or SIPC |
| Qualified custodian (for a fund or ETF) | The custodian, for the fund | Advisers Act custody rule; protection sits at fund level | You hold a share, not the coin |
Notice what the charter row does and does not do. It raises legal protection (federal supervision, enforced segregation, a fiduciary standard) without touching the technical-control column at all. Your keys are still held by someone else. For most of the custody spectrum, the real-world risk you keep is not the headline one. Self-custody’s danger is almost never the cryptography; it is the human, which is why phishing, not broken math, drains the most self-custodied wallets. The chartered custodian’s danger is not that it will abscond; it is that its private insurance is narrow and its failure, however unlikely, lands you in a bankruptcy court rather than a deposit-insurance line.
What a Charter Guarantees, and What It Does Not
Spell out the guarantees, because the marketing blurs them. A national trust bank charter gets you federal supervision by the OCC, a legal requirement to segregate client assets from the firm’s own, fiduciary duties that a court can enforce, and examination cycles that an ordinary exchange never faces. Those are real. In a bankruptcy, segregated client assets held in trust are much harder for the custodian’s creditors to reach than coins sitting in a general omnibus account under an exchange’s terms of service.
Now the limits, which are just as important. The Federal Deposit Insurance Corporation does not insure crypto, because crypto is not a deposit; the Securities Investor Protection Corporation does not cover it either, because a coin is not a registered security sitting in a brokerage account (SIPC). So the “insurance” a chartered custodian advertises is almost always a private crime or specie policy from a market like Lloyd’s, which pays for theft up to a sub-limit and does nothing for the firm’s own insolvency. When you read a custodian’s insurance headline, read the sub-limit and the exclusions, not the big round number.
A charter plus good compliance also does not make a custodian fraud-proof. Know-your-customer controls and licensing sit upstream of the vault, and criminals route around them, which is a running theme in how large-scale crypto fraud actually works in 2026, where the weak point is rarely the custodian’s cold storage and usually the human being tricked into sending funds in the first place. The charter raises the floor under a custodian. It does not promise a ceiling, and it does not reach past the custodian’s walls to protect a user who was socially engineered on the other side of the screen.
The Belshe Doctrine: Custody Should Not Sit Inside an Exchange
If one argument runs under the whole charter race, it is that holding coins and trading them are different jobs that should live in different companies. Mike Belshe, the chief executive of BitGo, has made the point as bluntly as anyone. “Exchanges have never held custody, never, of anything,” he told an interviewer in October 2026, arguing that execution, brokerage and custody belong in separate businesses because a custody failure at a trading venue takes the whole market down with it (The Block). His reference point is the most expensive lesson the industry has, the collapse of FTX, where customer coins and the trading desk’s risk sat in the same pocket until there was no pocket left.
The irony is that Belshe’s own firm spent 2026 complicating the thesis, launching a markets division that puts trading next to its chartered custody arm. Galaxy’s charter application does the same thing on purpose, bundling custody, trading and staking under one license. The charter regime, in other words, does not resolve the custody-versus-trading question; it formalizes both answers at the same time, with the segregation rules meant to keep the two functions from bleeding into each other inside a single chartered entity. For a retail user comparing venues, the practical read is simple: a drawdown is exactly when a trading venue’s custody promises get stress-tested, and the big exchanges differ more in how they hold and segregate coins than their near-identical marketing suggests.
The Bankruptcy Test: Which Law Decides Whose Coins They Are
The reason the fiduciary and segregation distinction matters is that it only gets tested when a holder fails, and the record on that is unambiguous. When Celsius went bankrupt, a federal judge ruled in January 2023 that assets in its interest-bearing Earn accounts belonged to the estate, not to the users, because the terms of service said so; Earn depositors became general unsecured creditors, while customers who had used Celsius’s separate Custody product were treated differently (CoinDesk). The lesson was blunt: the label on the account, not the fact that your coins once existed, decides who owns them when the music stops.
FTX taught the companion lesson. Creditors there are being repaid, cumulatively above the dollar value of their claims, but those claims were dollarized at the petition-date prices of November 2022, when Bitcoin sat near sixteen thousand dollars. A customer owed one Bitcoin got the 2022 dollar value of one Bitcoin, not a 2026 Bitcoin, and so missed the entire rally that followed. A chartered trust bank that holds your specific coins in a segregated fiduciary account is designed to avoid both outcomes: the coins are your property rather than the estate’s, and you are owed the coins rather than their stale dollar value. That is the single strongest practical argument for the charter, and it deserves to be stated plainly rather than buried under the licensing drama.
The catch is that the protection only holds if the coins are actually in a segregated trust account and not, through some product toggle, in a yield program or a pooled balance governed by different fine print. The account type does the legal work. A charter makes the segregated option available and supervised; it does not force a user to choose it, and it does not override whatever agreement that user clicked through.
Main Street Banks Sue to Stop It
Not everyone welcomes the new entrants. In October 2026 the Independent Community Bankers of America, the trade group for thousands of small US banks, sued the OCC to block the charters (American Banker). Its argument is jurisdictional: the National Bank Act, the ICBA says, does not authorize the OCC to hand national trust charters to firms whose core business is not traditional fiduciary trust work, and doing so, its complaint warns, “creates a gaping hole in financial regulation.” The group also makes a competitive case, that letting crypto firms operate as trust banks could pull deposits and activity out of the regulated banking system and into institutions with no obligation to lend to local communities.
The ICBA is not alone; the American Bankers Association and the Bank Policy Institute have filed their own objections, and the fight has become one of the sharper turf wars in US finance. The trade group points out that the OCC has approved or conditionally approved roughly twenty-one trust banks during the current administration, a pace it frames as a rushed expansion of federal chartering power rather than the routine licensing the regulator describes. Whether a court agrees is now a live question, and it hangs over every pending application in the table above, Galaxy’s included.
The Comptroller’s Case
The OCC’s answer is that it is doing nothing new. Gould, the Comptroller, has argued that safekeeping is an ancient banking function and that moving it onto a blockchain does not change its legal character. “New entrants into the federal banking sector are good for consumers, the banking industry and the economy,” he said, and he has publicly played down the community banks’ systemic-risk warnings (Banking Dive). The regulator’s position is that a limited-purpose trust bank which takes no deposits and makes no loans is, if anything, less risky than a commercial bank, because it is not running maturity transformation or credit risk; it is holding assets and keeping records.
Both sides are partly right, which is what makes the case interesting rather than obvious. The OCC is correct that custody is old and that a no-deposit, no-loan trust bank is a narrow thing. The community banks are correct that stretching the trust-bank charter to cover stablecoin issuers and trading-heavy shops is a real expansion of what that charter has historically meant. The resolution will set the perimeter of US crypto custody for years, and it will do so through litigation rather than legislation, which is exactly how most of the 2026 crypto rulebook has been written, from spot ETFs approved through exemptions and listing standards rather than a clean statute to custody itself.
The Charter Race Does Not Fix the Choke Point
Here is a limit the charter boosters rarely mention. Even as the number of chartered custodians grows, the US spot Bitcoin ETF market remains heavily concentrated in one of them. As of early 2026, Coinbase custodied roughly eighty-four percent of US spot Bitcoin ETF assets, about seventy-seven billion dollars across most of the major funds. Adding names to the custodian directory does not automatically spread that exposure, because the issuers choose the custodian, and inertia favors the incumbent with the longest track record and the deepest operational bench.
There is a second-order point hiding here. The charter race was sold partly as a cure for concentration, on the theory that more chartered custodians means more places for issuers to spread their coins. But charters and mandates are different markets. A charter is a supervisory status; a custody mandate is a commercial decision an ETF issuer makes weighing cost, track record and operational depth. Handing out more of the former does not force a redistribution of the latter, and so far the incumbents have kept most of the mandates even as the roster of eligible custodians grew.
2026 did show the first cracks. Grayscale named Coinbase Prime as a second custodian alongside BitGo for a new product, and BlackRock added Anchorage as a backup for its flagship funds, though it said it had no immediate plans to move assets. Naming a second custodian is not the same as splitting the coins. So the charter race and the concentration problem are related but distinct: chartering ten custodians does nothing for systemic risk if the ETF complex keeps parking the coins at one. For an ordinary investor who holds a spot ETF rather than the underlying asset, the practical exposure is to that single custodian’s operational integrity, wrapped in a fund structure that itself was approved through a regulatory side door. The charter improves the individual custodian. It does not, by itself, deconcentrate the market.
Charter Versus Article 75: The US and EU Took Different Roads
It is worth stepping outside the US for a moment, because Europe answered the same question with a statute instead of a charter race. Under the EU’s Markets in Crypto-Assets regulation, a licensed crypto custodian must, by law, keep client assets segregated from its own, and if it loses client assets or the means of access to them, it is liable to the client up to the market value of what was lost. Client assets are shielded from the custodian’s creditors in insolvency as a matter of regulation, not as a matter of each customer’s terms of service.
The contrast is philosophical. The US approach makes a specific custodian prudentially sound through a bank charter and then leaves the ownership question to trust law and the fine print of each agreement. The EU approach writes the segregation and liability rules into the regulation itself and applies them to every licensed custodian, chartered bank or not. Neither is automatically safer. A US national trust bank is a heavier, more closely supervised institution than most EU custodians; the EU framework hands users a baseline legal right that Americans have to assemble from trust law, account labels and bankruptcy precedent. The 2026 charter race is, in effect, the US betting that stronger institutions beat stronger statutes.
For a US holder, the practical takeaway is not that one system wins but that the protections live in different places. In Europe, a baseline of segregation and custodian liability comes attached to the license, so a retail user gets it more or less automatically. In the US, the equivalent protection has to be assembled: pick a custodian with the right charter, make sure the coins sit in a segregated trust account rather than a pooled or yield product, and understand that the final backstop is bankruptcy law and the account agreement, not a regulation that promises to make you whole. The charter is a strong ingredient, but the user still has to cook the meal.
How to Read a Custodian in 2026
Strip away the charters and the lawsuits and a user is left with a short list of questions that actually determine safety. A charter answers some of them and leaves others open, so it helps to ask them directly rather than trusting a badge.
- Is it a fiduciary? Ask whether the custodian holds your assets as a fiduciary with segregation, or as a counterparty under terms of service. The label decides who owns the coins in a bankruptcy.
- What does the charter actually cover? A national trust bank charter means federal supervision and segregation, not deposit insurance. Confirm which legal entity in the group holds the coins.
- What does the insurance really insure? Private crime or specie coverage pays for theft up to a sub-limit; it does not cover the firm’s insolvency. Read the sub-limit, not the headline figure.
- Can you verify the reserves? A custodian that publishes proof of reserves and lets you check your own balance is proving assets, though rarely liabilities. No proof at all is a red flag.
- Is custody separate from trading? If the same company runs the order book and the vault, a trading blow-up can reach your coins, as FTX showed.
- How concentrated is it? If you hold an ETF, find out which custodian holds the fund’s coins and whether a single failure there would freeze your exposure.
The charter race is genuinely good news, as far as it goes. A federally supervised, fiduciary custodian that segregates client assets is a better home for coins than an unregulated exchange omnibus account, and 2026 produced more of them than any prior year. But a charter is an upgrade to one row of the custody table, not a rewrite of the whole thing. It does not put crypto under deposit insurance, it does not end concentration, and it does not settle the bankruptcy question for anyone who clicks through a terms-of-service agreement without reading the account label. Who holds your coins still matters, and in a week when Bitcoin can fall fifteen percent before lunch, it matters most.
Frequently Asked Questions
What is a crypto national trust bank charter?
It is a federal license from the Office of the Comptroller of the Currency that lets a firm hold and safekeep digital assets as a fiduciary. A national trust bank in this group does not take insured deposits or make loans; its job is custody, record-keeping and standing behind client assets under federal supervision.
Does an OCC charter mean my crypto is insured by the FDIC?
No. The FDIC insures bank deposits, and crypto is not a deposit, so FDIC coverage does not apply even at a chartered trust bank. The SIPC does not cover crypto either. Any insurance a chartered custodian carries is usually a private crime policy that pays for theft up to a limit and does nothing for the firm’s insolvency.
Which crypto firms have OCC trust bank charters in 2026?
Anchorage Digital Bank was first, in 2021. In December 2025 the OCC conditionally approved Circle, Ripple, Paxos, BitGo and Fidelity Digital Assets, and through 2026 several moved to full approval, including Ripple in October. Crypto.com received conditional approval, and Galaxy Digital applied for its own charter in late October 2026.
Why are community banks suing the OCC over crypto charters?
The Independent Community Bankers of America argues that the National Bank Act does not give the OCC authority to charter firms whose core business is not traditional fiduciary trust work, and that doing so could pull deposits and activity out of the regulated banking system. Its October 2026 lawsuit asks the courts to block the charters.
Is a chartered custodian safer than keeping crypto on an exchange?
For legal protection, usually yes, because a fiduciary custodian must segregate client assets and a court is more likely to treat them as your property than coins held under an exchange’s terms of service. But a charter does not make the custodian hack-proof or insolvency-proof, and self-custody remains the only model with no third party to fail at all.
Yuki Tanaka covers wallets, exchanges and custody for HOGE Wire.