On a Tuesday that will look unremarkable on any price chart, the Federal Reserve published a proposal that quietly converted stablecoin issuance into a licensed banking activity. Two percent capital on the first $20 billion outstanding. One and a half percent on the next slice. One percent above the $50 billion line. A hundred percent reserve backing against a whitelist of assets narrower than most Treasury money market funds. And a yield provision that presumes — presumes — that routing interest to holders through a partner is a prohibited form of payment.
The market read the headline as a capital charge. That is the wrong frame. The capital charge is the cover story. The yield presumption is the weapon, and it is aimed directly at the business model the industry spent 2024 and 2025 building.
I have spent the past eighteen months modeling reserve adequacy for institutional clients weighing stablecoin exposure. The arithmetic in this document is not novel to me. It is the arithmetic I already run. The Fed has simply published my spreadsheet as federal law — and drawn the line exactly where the money sits.
Three overlapping texts now define US stablecoin regulation, and reading any one of them in isolation will mislead you. The GENIUS Act supplies the statutory skeleton — the first comprehensive federal framework for payment stablecoins, and the source of the Fed's authority here. The OCC issued a parallel proposal covering national bank issuance. The Fed's document, drafted under GENIUS, governs a specific and consequential perimeter: insured state member banks acting through subsidiaries, and uninsured state-chartered issuers above $10 billion in outstanding tokens.
That perimeter is not an accident. It captures Circle. It captures the bank-adjacent issuers. It reaches toward the largest compliant dollar tokens and leaves the largest offshore issuer — the single biggest stablecoin by supply — essentially untouched on US soil. This is not a universal framework. It is a set of rules written for the balance sheets the Fed can actually see.
Two structural features matter before we reach the numbers. First, the reserve definition. Eligible reserves are limited to cash, Federal Reserve balances, demand deposits, Treasuries of 93 days or less, overnight repo and reverse repo, funds investing solely in those instruments — and, notably, tokenized versions of them. That last clause is the interesting one. Second, the rebuttal mechanism. Issuers may rebut the yield presumption in writing. The burden falls entirely on the issuer, which is a polite way of saying the presumption wins by default. Code is law, but man is the loophole — and this document is an attempt to close the loophole by moving the burden of proof.
Let me deconstruct this from first principles, because the industry's reaction suggests most readers are arguing about the wrong layer.
A stablecoin, economically, is a demand deposit that pays no explicit interest and settles on rails the traditional banking system does not control. That is the whole thing. Everything else — the token, the chain, the smart contract — is distribution. The Fed's proposal recognizes this and regulates it as such. Three pillars, three constraints, one intended outcome.
Pillar one: full reserve backing. The 100% reserve requirement, with a whitelist that excludes corporate debt, commercial paper, and mortgage-backed securities, mirrors the post-2022 consensus on high-quality liquid assets. This is uncontroversial and, frankly, overdue. The 2022 algorithmic collapse that erased roughly $40 billion in days was a natural experiment in what happens when reserves are a promise rather than an asset. The Fed is codifying the lesson. The 93-day Treasury cap is the same instrument class a money market fund holds — and that is the tell. The regulator is not inventing a new asset class; it is mapping an old one onto a new wrapper.
Pillar two: tiered capital. Here the math gets interesting. Two percent on the first $20 billion means an issuer at that scale must hold an additional $400 million in high-quality liquid assets against the reserve base itself. At a 4% Treasury yield, that is roughly $16 million a year in foregone interest — a real cost, but absorbable. In 2020, when I built a Python simulation to stress-test lending pools against a 50% ETH drawdown, the finding was the same shape: the cost of safety is small in normal conditions and enormous at the margin. The tiering is deliberate — 2%, then 1.5%, then 1% above $50 billion. Marginal capital falls as size grows. This is scale-economy logic applied to systemic risk, and its practical effect is consolidation. Small issuers carry the highest relative compliance load. Large issuers carry the lowest. The regulation is not neutral between competitors; it is engineered to reward the balance sheets that already exist.
The second capital requirement — 25% of the three-year average of non-reserve revenue — is the clause almost nobody discussed, and it deserves more attention. It targets issuers who try to escape the reserve cage by building lending or advisory businesses on the side. The message is precise: you cannot construct a fee-driven financial conglomerate around a stablecoin and treat the reserve as a loss leader. The Fed is capping the scope of the business, not just its leverage. This is a structural constraint dressed as a capital ratio.
Pillar three: the yield presumption. This is the load-bearing wall. The proposal presumes that payments from an issuer to a related party or third party — where that party then pays holders — constitute a prohibited interest payment. It covers service providers and white-label partners. It reaches arrangements where a brand issues a token, a partner distributes it, and the economics flow back through a side channel.
Read that carefully and you see what it kills. Yield-bearing stablecoins built on third-party revenue sharing. White-label tokens where the brand pays holders out of marketing budgets. Loyalty-style reward programs dressed as token balances. The proposal does not ban these by name; it presumes them illegal and makes the issuer prove otherwise. That is a far more effective instrument than a prohibition, because it shifts legal cost and uncertainty onto the very players least able to carry it.
And here the DeFi parallel is unavoidable. The interest rate models inside the largest lending protocols — the curves everyone treats as market-derived — are administrative parameters set by governance, not prices discovered by supply and demand. I said this in 2020 and it remains true: those rates are arbitrary. What the Fed is doing is applying the same administrative logic from the other direction. If the protocol sets yield by committee and the regulator sets yield by presumption, then the only genuinely market-determined yield in the stack has just been regulated out of existence. The presumption does not discover the price of money. It decides that the price should not be paid.
Now the macro overlay that the crypto-native crowd keeps missing. This is not a crypto story. It is a monetary policy story. A deposit that pays yield competes with bank deposits and with Treasury bills, and the Fed's mandate touches both. The presumption is not consumer protection. It is deposit-substitution management. When I mapped correlation matrices between stablecoin supply growth and money-market fund flows across the 2022 contraction, the coefficient was uncomfortable — stablecoin growth tracked precisely the windows when yield-seeking capital had nowhere else to go. Regulators noticed the same pattern. This document is the response, and it is a response calibrated to the 2010 money market fund reforms, which similarly traded yield for stability after a run exposed the fragility underneath.
The one silver lining, and it is real: tokenized Treasury funds are named as eligible reserves. That quietly blesses on-chain government money as a reserve asset and opens a compliant path for the RWA infrastructure that spent two years searching for institutional legitimacy. The reserve side gets a boost at the exact moment the yield side gets gutted.
The consensus interpretation is that this framework strengthens the compliant dollar — Circle wins, the bank issuers win, the offshore challenger gets marginalized, and the market consolidates toward safety. I think that is half right and dangerously incomplete.
The blind spot is the perimeter itself. Because the Fed's reach stops at insured state member banks and large uninsured state issuers, it creates a clean regulatory arbitrage at the edges. A small uninsured issuer under $10 billion sits in a lighter regime. An offshore issuer is untouched entirely. You cannot regulate a dollar-denominated instrument globally by regulating the institutions that happen to be chartered in your jurisdiction. The dollar is the product; the charter is the wrapper. Anyone who wants to issue a yield-bearing synthetic dollar will simply do it one border away, and the US will have exported the risk while importing none of the oversight. This is the same structural failure that allowed cumulative cross-chain bridge losses to exceed $2.5 billion while the industry kept building more bridges. Regulating the entity does not regulate the activity.
The second blind spot is the timeline. The effective date is January 18, 2027, or 120 days after the final rule — whichever comes first. The Clarity Act, which would settle the reward question in statute, is stalled in the Senate precisely because banks pushed to restrict those rewards. So we have a regulator moving ahead of the legislature, writing a presumption into a comment-period document while the legislative question stays open. If Clarity later permits what the Fed has presumed illegal, pillar three collapses into litigation. The next eighteen months are not implementation. They are a legal war fought through comment letters.
Here is how I am positioning. The compliant dollar becomes a utility — thin margins, high volume, bank-adjacent, and eventually a bridge product that traditional institutions use to enter digital assets without ever admitting they hold crypto. The yield-bearing dollar migrates offshore and onto chains the Fed cannot reach, and it will keep the yield, because someone always will. The bet that matters is not which stablecoin wins. It is whether the line between them holds when the first large issuer stands up a subsidiary in a friendly jurisdiction and calls it compliance.
Code is law. Whoever writes the charter decides where the code runs.