I keep coming back to the same question whenever a regulator drops a thick packet on payment tokens: does this actually change how a dollar on-chain is supposed to sit in the real world, or is it just more paperwork around a product people already treat as cash? The latest Federal Reserve drafts under the GENIUS Act land closer to the first answer than the second. They talk about full backing, permitted reserve assets, capital for credit and operational risk, and a formal path for an insured state member bank that wants a subsidiary to mint payment stablecoins. That mix is dry on the page. It is not dry if you hold the tokens, work at a bank considering a consortium, or simply care whether a so-called dollar can be redeemed when markets get ugly.
What The Fed Put On The Table
Two proposals arrived together. One would set operating rules for Fed-supervised issuers and for firms that hold the assets behind those tokens. The other would tell an insured state member bank how to ask for permission so a subsidiary can issue payment stablecoins. Public comments would run for 60 days after the notices show up in the Federal Register. There is no neat calendar date in the announcement itself, which is the sort of detail that makes compliance desks twitch.
I’ve found that people skim the headline and assume every U.S. bank can now print a branded dollar token by Christmas. That is not what these drafts do. They put flesh on parts of the statute that already exist. They remain open to rewrite. And they do not turn a payment stablecoin into an insured deposit. That last point is worth repeating slowly. Full reserve backing is not deposit insurance. Holders should not confuse the two, even if marketing teams will try.
Full backing means assets against every token issued, not a thinner cushion hoping redemptions stay polite.
Full Backing Is The Core Demand
Under the first draft, a Fed-supervised issuer would have to hold permissible assets that fully back outstanding payment stablecoins. Short-term U.S. Treasury bills sit in the example pile, along with certain other high-quality liquid instruments. The idea is simple even if the legal text will not be. If you issued a hundred tokens, you do not get to warehouse eighty dollars of good collateral and call the rest a model. You hold the stack.
That sounds obvious until you remember how earlier market stress treated “mostly reserved” products. When redemptions bunch, haircuts and liquidity gaps stop being theoretical. Full backing is the Fed’s way of saying the float is not a piggy bank. In my experience, that single sentence does more work than pages of risk-committee jargon.
The same proposal would standardize capital requirements aimed at credit and operational risk tied to the activity. Separate risk-management standards would cover how the issuer actually runs the shop. Firms supervised by the Fed that safeguard backing assets would get their own rule set. Custody is not a side job in this frame. It is part of the product.
- Permitted reserves would need to match tokens in circulation.
- Capital would speak to credit and operational exposures, not just market slogans.
- Safekeeping firms under Fed watch would face dedicated expectations.
- Payment stablecoins would still sit outside insured deposit status.
What Banks Could Do Besides A Subsidiary
Buried near the reserve language is a quieter piece. The draft would also clarify which stablecoin and related activities Fed-supervised banks may undertake without treating every idea as a brand-new charter play. That matters. A bank might want to hold reserves, process redemptions, or support distribution without becoming the face of the token. Clearer activity lists reduce the “call your lawyer twice” tax that has slowed otherwise ordinary operations.
Perhaps the most interesting aspect is how ordinary that clarification sounds and how rare it has been. Banks have spent years asking which box a token belongs in. A proposed regulatory basis for activities beyond a subsidiary application is not glamorous. It is how products actually ship.
The Bank Application Path, Step By Step
The second proposal is narrower. It applies to insured state member banks that want Fed approval for a subsidiary to issue payment stablecoins. The bank files with its appropriate Reserve Bank. The bank, not the proposed subsidiary, is the applicant. That allocation of responsibility is not a clerical quirk. It puts the parent’s balance sheet, governance, and reputation on the form.
Applicants would send a business plan, financial information, and whatever else the Fed needs to judge the operation. The filing would describe the plan, state the approval sought, and explain why it should be granted under the statute’s factors. Hearings, appeals, and final decisions get procedures of their own. None of this is mysterious. It is bank supervision wearing a token costume.
Timing is where operators should sit up. The Fed would say within 30 days whether a filing is substantially complete and what is missing if it is not. Once the file is substantially complete, the law gives the agency 120 days to decide. If it does not decide in that window, a complete application is deemed approved. That deemed-approval clause is the sleeper. Agencies hate clocks they cannot stretch. Banks love clocks they can plan around. Both instincts will show up in comment letters.
| Stage | Clock | What It Means |
| Completeness review | 30 days | File is whole, or gaps are listed |
| Decision window | 120 days after complete | Approve, deny, or deemed approve |
| Material change | New date possible | Plan, ownership, or finances shift |
| Comment window | 60 days after Register | Public can still reshape the draft |
When A Plan Change Restarts The Clock
A substantial change to the proposed issuer’s business plan, ownership, or financial condition may require more information and a new submission date. That is the sentence consortium lawyers will highlight in yellow. Group projects change. Partners join. Capital stacks get redrawn. If every tweak resets completeness, the 120-day promise becomes a moving target.
The draft also asks for comment on applications involving several banks in a stablecoin consortium. Could one filing cover participating insured state member banks in some cases? That is not academic. A large group of institutions has already said it wants a dollar token and intends to meet applicable statutory requirements, with a first-half 2027 target floating in public remarks. That announcement does not prove they will use this exact Fed route. It does prove multi-bank structures are no longer a thought experiment.
I’ve sat through enough joint-venture decks to know the hard part is rarely the white paper. It is who owns the key, who holds the T-bills, who answers the examiner, and who eats the first operational miss. A single filing versus a stack of parallel filings will decide months of calendar time.
How This Sits Next To Other Agencies
The Fed is not writing in a vacuum. Treasury has already floated definitions for when payment stablecoins are issued, offered, or sold in the United States. Those questions police the border of licensing and distribution. The Fed drafts police issuers it supervises and applications from insured state member banks. Different doors, same house.
The national bank supervisor has been building a separate framework for issuers under its own authority, covering reserves, redemptions, custody, supervision, and applications, with industry feedback already in the mix and a late-year target discussed in public remarks. That timetable does not set a finish line for the Fed packets released now. Anyone hoping for one clean federal rulebook will wait longer than they like.
Statute-watchers will recall an expected effective date in mid-January 2027 for the main issuer restrictions, plus a possible earlier start 120 days after responsible federal regulators finish implementing rules. Agencies already missed an earlier completion deadline in the law. Several proposals now sit at different stages. That stagger is messy. It is also how U.S. financial rulemaking usually looks when more than one charter type is in play.
A payment stablecoin can be fully reserved and still fail a holder’s mental model if people think insurance is attached.
Why Reserve Quality Beats Reserve Theater
Short-term Treasuries show up as examples for a reason. They settle in a market deep enough that a large redemption does not have to become a fire sale of odd credits. “Certain other” high-quality liquid assets will be the fight in the comments. Every extra category is a bid from someone who wants yield, flexibility, or both. Every extra category is also a place where liquidity can vanish on a bad Monday.
In my view, the boring portfolio is the point. Payment tokens that want to behave like cash should not need a credit committee to explain the weekend. If an issuer needs a slide titled “why this asset is basically cash,” it probably is not.
- Map every token to an eligible reserve item, not a blended average.
- Test redemption under bunched outflows, not average daily volume.
- Separate custody risk from issuer operational risk in the capital math.
- Write the holder disclosure as if no one reads footnotes.
Capital, Operations, And The Quiet Custody Layer
Capital rules for this activity will look modest next to a trading book and enormous next to a software shop that once treated tokens as a side feature. Credit risk on permitted reserves should be small if the asset list stays tight. Operational risk will not be small. Smart-contract bugs, key ceremonies, reconciliation breaks, vendor outages, and fat-finger mints are the real animals in this zoo.
Risk-management standards will live or die on whether examiners can follow the flow of funds from mint to reserve to redeem without a translator. If the ledger and the custody account cannot be tied on the same day, the model is a story. Stories do not survive a run.
Firms that hold backing assets under Fed supervision would get dedicated rules. That is overdue. A token can be perfectly designed and still wobble if the custodian’s controls are theater. I would rather see dull, testable safekeeping than a prettier dashboard.
What This Means For A Token Holder In The United States
If your token is issued by a firm inside the Fed’s perimeter, the reserve draft is about the stuff behind the ticker, not a government wrap on your wallet. You still need to know who issues, who holds the bills, how fast redemption works, and what happens if the issuer stumbles. Those questions existed before this packet. They are now more likely to have standardized answers.
If your token sits outside that perimeter, these particular drafts may not be your rulebook. Distribution definitions from other agencies may still catch the offer or sale. Cross-border products love that gray zone. Holders should not assume a U.S.-looking dollar ticker automatically lives under this Fed text.
Is that frustrating? Yes. Is it surprising? Not if you have watched charters multiply for a decade.
Consortium Politics Will Test The Application Draft
Multi-bank tokens sound efficient until governance starts. Who appoints the subsidiary board. Who can halt minting. Who funds a capital call. Who talks to the Reserve Bank when the file is incomplete. A single application covering several insured state member banks could cut duplication. It could also hide disagreements until the completeness letter arrives.
A material change rule will punish sloppy syndicates. If ownership shifts after filing, the clock may move. That is healthy. It is also a reason groups should freeze the cap table before they hit send, even if the press release wants more logos.
Application hygiene, plain version: Freeze the plan before filing Name the parent as applicant Document reserves as if examiners will sample them Assume a plan change can reset the date
The Comment Period Is Not Ceremonial
Sixty days after Federal Register publication is short for a product that touches payments, custody, and bank subsidiaries. Interested banks, issuers, and the broader public can file through the Fed’s online proposal system, mail, or email, tagged to the relevant docket. That sentence will look like boilerplate. It is not. The deemed-approval clock, the definition of a substantial change, the consortium filing question, and the edges of permitted reserves will all be argued here.
I’ve found comment letters work when they attach an operational scenario, not a sermon. “Here is what happens if two banks in a group change ownership on day 40” beats “please be innovation-friendly.” Examiners already know the slogan. They need the failure mode.
A Realistic Timeline Without Magical Thinking
Register publication, then 60 days of comments, then revision, then a final rule, then implementation. Layer that on a statutory effective date that can also hitch to finished federal rules. Add parallel work at other agencies. The result is not a single big-bang morning when every lawful payment stablecoin appears. It is a staggered opening, charter by charter.
A first-half 2027 product target from a bank group can still be real if the application path is clean and the reserve stack is dull. It can slip if completeness fights eat the winter. Anyone selling certainty before the comments close is selling something else.
Where I Think The Fight Will Actually Be
Not on the slogan of full backing. That fight is mostly over in respectable rooms. The fight will be which assets count, how capital is calibrated for operational misses, whether one consortium form is allowed, and how brutally a plan change resets the 120-day clock. Those are plumber questions. Plumber questions decide whether a product ships.
There is also a cultural fight hiding in the insurance disclaimer. If banks market a token next to a checking account, holders will hear “safe” even when the legal text says otherwise. Disclosure can be accurate and still lose to a brand color. That is not the Fed’s drafting problem alone. It will become everyone’s problem the first time a redemption queue hits social media.
Practical Takeaways Without The Fog
If you issue or want to issue under Fed supervision, start mapping reserves to tokens one-for-one and write the operational risk story as if an examiner will pull a sample tomorrow. If you are an insured state member bank eyeing a subsidiary, treat the parent as the applicant and freeze the plan before the completeness clock starts. If you hold tokens, ask who backs them and how redemption works, then remember that backing is not insurance.
If you sit in a multi-bank group, decide now whether you need one filing or many, and decide who can change the business plan without dragging everyone back to day one. If you write comments, pick one operational knot and pull. Scattershot letters get filed. Tight letters get read.
- Full reserve backing is the center of gravity for supervised issuers.
- Capital and risk controls will follow the activity, including safekeeping.
- Bank subsidiaries enter through the parent’s application, not a side door.
- Thirty days tests completeness; 120 days tests the agency; deemed approval lurks.
- Consortium structures and plan changes are the live wires in the second draft.
None of this makes payment stablecoins boring overnight. It does make the respectable version of the product look more like cash-management infrastructure and less like a slogan with a smart contract taped on. That shift is overdue. It will still be argued, line by line, once the Register clock starts. And that, more than any headline about “banks entering crypto,” is the part worth watching with a pencil in hand.
