There are two roads to tokenized money, and bankers shouldn't ignore either one

The US and Europe have made opposite bets on who should issue tokenized money

All money is a promise. Your bank balance, your PayPal account, the USDC in your wallet. Each one is an IOU from some institution, redeemable (in theory) for something else. The quality of any form of money depends on the reliability of the debtor and the legal infrastructure around it. That’s true whether the debtor is JPMorgan, Circle, or the European Central Bank.

This matters right now because the US and Europe have made opposite bets on who should issue tokenized money and what legal protections should surround it. Europe is building a digital euro. The US has said, in about as many words, “absolutely not” and has instead written rules for private stablecoins. The stablecoin market crossed $300 billion in late 2025. Nine crypto native firms have received OCC national trust bank charters (most still conditional), with more applications in the pipeline. Both paths will reshape competition in banking. Neither is obviously right. And I think the structural risks in each approach are poorly understood even by most people in the industry.

The European approach is upgrade the plumbing, preserve the banks

The ECB is developing a retail digital euro. The preparation phase ran through October 2025, with enabling legislation expected in 2026, a pilot around mid 2027, and possible issuance by 2029. Separately, the ECB has launched wholesale CBDC projects (Pontes and Appia) for interbank settlement. On top of this infrastructure, banks would issue tokenized deposits, so your commercial bank money represented as tokens on a distributed ledger, with the same deposit insurance and regulation that applies today.

The EBA published a report on tokenized deposits in December 2024 that should worry anyone banking on this timeline. Out of 85 surveyed banks, one had a live tokenized deposit product in the EEA. One. About 17% expected to engage within two years. The rest were waiting.

The architecture preserves the two tier banking system while upgrading it to run on programmable infrastructure. European bankers keep their role, but they need to build. If they don’t offer tokenized deposits, licensed stablecoin issuers under MiCA will. Circle obtained an EMI license from France’s ACPR in July 2024, allowing it to issue USDC and EURC across the EU. It’s pretty clear that competition is already in the market.

The American approach is let the private sector figure it out

On January 23, 2025, Trump signed an executive order prohibiting federal agencies from developing or promoting a CBDC. The concern was surveillance. Congress broadly agreed. Instead, the US built rules for private stablecoins. The GENIUS Act became law on July 18, 2025, permitting three types of issuers. 1) subsidiaries of insured depository institutions, 2) OCC supervised non banks, and 3) state chartered issuers under $10 billion in circulation. Implementing regulations are due by July 2026.

Then came the charters. Since early 2025 the OCC has been processing a wave of national trust bank applications from crypto-native firms, and the approvals have been fast. Anchorage Digital, which received its conditional charter in 2021, is now fully operational as the only nationally chartered crypto bank. Erebor Bank received conditional approval in October 2025 and final charter approval in February 2026. On December 12, 2025, the OCC approved five more at once: Ripple and Circle (as First National Digital Currency Bank) as de novo charters, plus BitGo, Fidelity Digital Assets, and Paxos converting from state trust companies. Bridge (the Stripe subsidiary) got conditional approval in February 2026. Crypto.com followed shortly after.

These are trust bank charters, not full commercial bank licenses. The firms can’t take deposits, can’t access FDIC insurance, can’t borrow from the discount window. What they get is federal supervisory legitimacy under the OCC, preemption of state by state money transmitter licensing, and, because national banks are required to be Federal Reserve members, eligibility for Fed master accounts. The Bank Policy Institute, representing traditional banks, has argued that these novel charter institutions shouldn’t automatically qualify as “depository institutions” under the Federal Reserve Act. But the legal default favors them as nationally chartered banks.

Why convenience will win and what that means for deposits

There’s an old idea in monetary economics that bad money drives out good, people will hoard gold coins and spend debased ones. In a digital payments world I think the dynamic runs the other way. Consumers optimize for convenience. They’ll move to whatever system is cheapest and most frictionless and stop caring whether the money inside it is safe. If you think about PayPal, Venmo, M-PESA, Alipay etc, there are hundreds of millions of people already hold monetary IOUs from non bank companies with weak regulatory protections. Why? Well, because the payments work and nobody reads the fine print.

That’s really the dynamic the US has now turbocharged. The GENIUS Act creates a regulated stablecoin market where newly chartered trust banks will issue, custody, and settle tokenized dollars that compete directly with bank deposits for payments and programmable finance. If stablecoins offer faster and cheaper payments, and running on 24/7 blockchain rails, then we can expect some portion of deposits will migrate. The OCC charter wave means the firms capturing those flows won’t be startups with uncertain regulatory status. They’ll be federally supervised institutions with national charters.

In Europe, the architecture is explicitly designed to prevent this. The digital euro provides the settlement layer; banks provide the deposit layer. MiCA requires overcollateralization of stablecoins, with custody asset values exceeding outstanding token values. Banks keep issuing money. Stablecoins compete on the margin.

In the US, the structural protection for bank deposits doesn’t exist. Which means the strategic question for US banks isn’t whether stablecoins will compete with them. It’s really how much market share they’ll lose, and how fast.

What happens when a stablecoin issuer fails

Bank deposits survive bank failures. Deposit insurance covers the first $250,000. The Fed acts as lender of last resort. Prudential regulation limits risk-taking. This is the infrastructure that makes bank money reliable in a crisis.

Stablecoins, even GENIUS Act compliant stablecoins issued by OCC trust banks, have none of this. No FDIC insurance. No discount window. No lender of last resort. Reserve requirements and federal oversight, yes. But what happens in bankruptcy?

When a company enters bankruptcy, the automatic stay freezes everything. Creditors can’t withdraw. Under the pari passu rule, stablecoin holders get lumped in with every other unsecured creditor. The Celsius bankruptcy showed what this looks like: in January 2023 a court ruled that crypto in Celsius’s Earn accounts was property of the bankruptcy estate, not the depositors’. About $4.2 billion in accounts became unsecured claims. The court applied the logic of the 1848 English case Foley v. Hill, which holds that money deposited with a bank belongs to the bank.

Bankers might dismiss Celsius as a sketchy crypto lender. But the legal principle applies broadly. A GENIUS Act trust bank issuing stablecoins with full reserves could face the same automatic stay in bankruptcy. The GENIUS Act’s reserve and custody requirements may function as structural separation that keeps customer assets outside the estate, or they may not. That question hasn’t been tested in court. Japan offers a useful comparison: after Mt. Gox, Japan mandated strict asset segregation for crypto firms. When FTX collapsed in November 2022, Japanese customers got their money back in three months while US customers were still in line.

Whether the GENIUS Act’s protections are strong enough to produce a Japan like outcome or a Celsius like one is something we won’t know until there’s a failure. And with over $260 billion in USDC and USDT alone, the stakes when that test comes will be large.

The SVB problem, but multiplied across borders

In March 2023, Circle had about $3.3 billion in reserves at Silicon Valley Bank. When SVB failed, USDC briefly traded at $0.8774. The depeg reversed only because the FDIC covered all SVB deposits, including uninsured ones. The stablecoin was only as safe as the bank holding its reserves, and most USDC holders had no idea which bank that was.

This is the correspondent banking fragility at the heart of stablecoin architecture. It’s worth considering there’s a bit of a multi issuance problem. USDC is one brand but it’s issued by different legal entities in different jurisdictions, under MiCA in Europe and the GENIUS Act in the US. Which entity’s reserves back your specific tokens? If you need to redeem, who are you redeeming against?

The equivalence frameworks between MiCA and the GENIUS Act, the mechanisms for cross border regulatory recognition, are underdeveloped. For banks that custody stablecoins or lend against them as collateral, this is basically a credit risk problem dressed up as a technology question.

AML should be keeping compliance up at night

Chainalysis reported $154 billion in illicit crypto volume in 2025, with stablecoins accounting for 84% of it, up from 63% the prior year. The specific instrument both jurisdictions are building their regulatory frameworks around is also the preferred vehicle for illicit finance. Banks touching this ecosystem need blockchain analytics capabilities most of them haven’t invested in yet.

Two paths to Fed access

The OCC trust bank charters carry Fed access as part of the package. As nationally chartered banks, these institutions are required to be Federal Reserve members and are generally eligible for master accounts, giving them direct access to Fedwire. That’s not the skinny master account, that’s the front door.

The skinny master account is a separate pathway, designed for state chartered entities that don’t have an OCC charter. In October 2025, Fed Governor Christopher Waller proposed limited purpose accounts with settlement access but no discount window, no interest on reserves, and probable balance caps, restricted to firms with a state SPDI or similar charter. The Fed is targeting a broader Q4 2026 rollout.

The distinction matters because of what happened with Kraken and Custodia, two Wyoming SPDIs that went after the same thing and got opposite results. Custodia applied for a master account in 2020 and was denied by the Kansas City Fed in 2023, during a period when Tier 3 access (for eligible but federally uninsured institutions) was effectively unobtainable. The district court upheld the denial. The 10th Circuit affirmed in October 2025. Custodia is seeking en banc review.

Kraken Financial, also a Wyoming SPDI, applied in the same period but got its answer in a different political environment. On March 4, 2026, the Kansas City Fed approved a limited purpose master account for Kraken, making it the first crypto firm in history with direct Fedwire access. The approval is deliberately constrained: no interest, no emergency lending, one year initial term, and it serves as a pilot for the broader skinny account framework. The Fed classified Kraken as Tier 3, only the third institution at that tier ever approved.

So there are now two routes to Fed settlement for crypto native firms. OCC trust banks get it through membership. State chartered SPDIs can get it through the skinny account pathway that Kraken just proved works. Either way, these firms no longer depend on commercial banks for access to the payment system. That removes the correspondent banking vulnerability exposed in the SVB episode. It also removes a revenue stream for bank with the fees that correspondent banks earn for providing that access. For the largest custody and clearing banks, the competitive implications of a dozen newly chartered institutions with direct Fed access should be getting attention at the board level.

What’s still missing

A properly designed tokenized money system needs three things. First, a regulatory charter for payment issuers with genuine structural separation of customer funds and clear resolution procedures. Second, open access to central bank settlement. Third, governance: an institutional framework for how stablecoin dollars, tokenized deposits, and traditional bank money coexist as all three scale.

The US is making real progress on the first two. The GENIUS Act provides the charter; the OCC is filling it with institutions. OCC trust banks get Fed access through membership; state chartered SPDIs now have the Kraken precedent and the skinny account rollout targeting Q4 2026. Governance remains completely absent. Nobody is really coordinating how these parallel monetary systems will interoperate.

The EU has more of the architecture on paper. The digital euro provides settlement. MiCA provides the charter. The ECB provides governance. But with one live tokenized deposit and most banks waiting, the architecture is largely theoretical at this point.

The BIS’s Project Agora, with seven central banks and 40+ financial institutions, is testing tokenized commercial bank deposits settling against wholesale CBDC. It’s the closest thing to a complete working model. But it’s also a lab experiment.

The cost of financial intermediation in the US has been stuck at 1.5 to 2 percent for over a hundred years. Tokenized money is a bet that technology finally breaks that. The two roads the US and Europe are taking do not look like they are going to be converging anytime soon, and both involve regulatory path dependency: charters granted, infrastructure built, relationships locked in. The decisions being made in 2026 feel like they will be very expensive to reverse by 2030.

---

Sources

Awrey, Dan. Beyond Banks: The Future of Money. Princeton University Press, 2024.

Chainalysis 2026 Crypto Crime Report. Chainalysis, 2026.

European Banking Authority Report on Tokenised Deposits. EBA, December 2024.

Ferretti, Federico, et al. The Tokenized Economy. Giappichelli, 2026.

GENIUS Act — Congressional Research Service Overview. Signed into law July 18, 2025.

In re Celsius Network LLC, No. 22-10964 (Bankr. S.D.N.Y. 2023).

Custodia Bank v. Federal Reserve Board of Governors, No. 24-8024 (10th Cir. 2025).

Federal Reserve Governor Waller: “Skinny” Master Account Proposal. Mayer Brown, October 2025.

Federal Reserve targets Q4 2026 for skinny master account rollout. Troutman Pepper, December 2025.

Kraken becomes first digital asset bank to receive a Federal Reserve master account. Kraken Financial, March 4, 2026.

Kraken receives Fed master account, in a first for crypto. Banking Dive, March 2026.

OCC Announces Conditional Approvals for Five National Trust Bank Charter Applications. OCC, December 12, 2025.

Erebor Bank approved for conditional federal bank charter. CoinDesk, October 2025.

Crypto.com receives conditional OCC national trust bank charter. Banking Dive, 2026.

Stablecoin market tops $317 billion. MEXC, January 2026.

Philippon, Thomas. “Has the US Finance Industry Become Less Efficient?” American Economic Review 105, no. 4 (2015): 1408–1438.

Regulation (EU) 2023/1114 — Markets in Crypto-Assets (MiCA).

Trump, Donald J. “Strengthening American Leadership in Digital Financial Technology.” Executive Order, January 23, 2025.