The most consequential number in the Federal Reserve's latest stablecoin draft is not 1:1. That ratio is table stakes — a consensus floor now shared by the EU's MiCA, Hong Kong's stablecoin ordinance, and Singapore's MAS framework. The number that matters is two.
Two days to redeem. Two days to turn a reserve portfolio into settled cash and return it to a holder who wants out. Here is the anomaly that should stop any engineer cold: since May 2024, US Treasury trades on a T+1 settlement cycle. The Fed is proposing a redemption commitment that is looser than the settlement of the very assets most likely to back the token — but only if every link holds. Cash pending. Custodian ready. Counterparty not a bank caught in its own run.
Based on my own audit work across 2024 and 2025, that "only if" is precisely where stablecoin economics break. Code does not lie, but it does hide.
Context
The draft does three things. It mandates full 1:1 backing by eligible assets. It requires redemption within two business days. It compels weekly reporting. Each mandate is deceptively simple at headline level and expands into pages of definition once drafted for implementation.
The publishing authority matters: this comes from the Fed, not the SEC or the OCC, which signals the rule is drawn around bank-affiliated issuers and their holding companies. The phrasing "refining issuance rules" implies an existing parent statute already exists — meaning we are not watching the birth of law. We are watching the operational spec.
That distinction is the entire story. A parent statute sets intent. An implementation rule sets behavior, and behavior is where capital survives or dies.
Core
Start with duration. A 1:1 reserve mandate is a ratio ceiling; a two-day redemption mandate is a duration ceiling. They are not the same constraint, and the second is far harder. Consider a reserve built from short-dated T-bills and overnight reverse repos — the Circle-style stack. T+1 settlement of a bill already held clears inside the window. The math works. Now tilt the portfolio toward mid-duration notes or offshore custody accounts, and those same two days become a forced-sale clock. To raise cash faster than the market settles, you sell at a discount. Run enough issuers through a coordinated withdrawal, and the discount becomes the exact loss that 1:1 was designed to prevent.
This is why I treat the ratio as the easy rule and the window as the real one. A ratio is a snapshot. A redemption term is a promise about behavior under stress — and a guarantee is a promise to be liquid on the worst day, not the average day. Regulation written around average days fails on the tail, and the tail is the only region that matters.
Weekly reporting compounds the difficulty. To disclose reserve composition every week to a supervisor, an issuer needs near-real-time reconciliation between the chain and the custodial ledger. Most issuers today rely on monthly attestation: a point-in-time statement that the books matched on a given date. Weekly cadence is a different engineering problem. It requires a programmatic feed to the custodian, automated position mapping, and an auditor who can operate on that rhythm. The largest licensed issuers have built parts of this. The rest have not.
So the rule does not primarily regulate stablecoins. It regulates who can afford to run one. The trade is compliance for yield: shrink reserve duration, accept lower interest income, buy a license. That trade is brutally asymmetric by size. A trillion-dollar issuer absorbs the cost as a line item. A hundred-million-dollar issuer gets squeezed toward consolidation or offshore.
One more mechanical note from the audit side. The two-day clock has no defined stop. Does the timer start at the on-chain request, at KYC confirmation, or at fiat settlement? Each definition moves the real deadline by up to a day. The best audit is the one you never see, and the sharpest compliance risk here is a definitional gap, not an enforcement one.
Contrarian
Now the blind spot. The market is reading this as a crackdown on opaque issuers. The contrarian reading is that the crackdown is optional.
Two-day redemption and weekly reporting bind only entities that choose to serve US customers through the supervised banking perimeter. An offshore issuer can keep a longer duration, disclose monthly, and simply restrict US access. The rule's actual teeth therefore live in its extraterritorial clauses and reserve-custody requirements — the parts the headline never touched. If the draft constrains where a dollar reserve can sit rather than who can transact, it bites hard. If it does not, it manufactures a two-track market: a compliant, slower, lower-yield dollar onshore, and a faster, higher-yield, less-audited dollar elsewhere. That split is a feature of the perimeter, not a bug in the text.
The second blind spot is yield. A payment stablecoin with no profit expectation stays clear of Howey. A yield-bearing stablecoin that distributes reserve income drifts toward the securities question, and that is the real dividing line in this framework. If the final text restricts distribution of reserve yield, it does not merely tighten disclosure — it deletes a business model. The front-runners are already inside the block here: issuers are repricing toward short-duration reserves before the rule is final, which tells you the direction is priced and only the severity remains open.
Takeaway
Watch three lines over the next two quarters: the reserve-eligibility list, the definition of redemption — specifically whether atomic on-chain redemption counts against the two-day clock — and the extraterritorial language. Those three clauses decide whether this becomes a moat for licensed issuers or a partition of the dollar itself. The rule was never the risk. The unread clause is.
