Nobody Holds the Licence Yet: The Stablecoin Market Stalled at $320bn While Washington Missed Its Own Deadline — and the Wrappers Are Onshoring to the OCC, Not to Cayman

The Licence Nobody Holds

Count the entities legally authorised to issue a payment stablecoin in the United States under the GENIUS Act, one year after the President signed it. The number is zero.

Not "few." Not "a handful of early movers." Zero. As of 14 July 2026, every GENIUS implementing document in the record is a proposal, not a final rule: the OCC's is a Notice of Proposed Rulemaking, and FinCEN and OFAC issued a joint proposed rule in April. Because the statute's 120-day trigger runs only on final regulations, no clock has started, no requirement binds, and there is no approved permitted payment stablecoin issuer to name. Every stablecoin circulating today was issued under pre-GENIUS arrangements. [2] [10] [1]

Now the second fact, which is the one nobody wants. The market this statute was written for has stopped growing. The Federal Reserve put the aggregate stablecoin market at $317 billion as of April 6, 2026, with growth having "flatten[ed] out during the final quarter of 2025 and first quarter of 2026." [11] A month later the Fed's own Financial Stability Report said, flatly, "Stablecoin growth moderated," and put the level "currently $320 billion, concentrated among the two largest issuers." [12] Tether — the larger of those two — describes its own float in its BDO-attested quarterly report as having "remained broadly stable at scale." [14]

So the received story — that regulated stablecoin issuance is scaling fast, and that the legal wrappers for it are migrating to the Cayman Islands — is wrong twice, and not marginally. Nothing regulated has begun issuing. Nothing is scaling. And the wrappers are moving in precisely the opposite direction from Cayman: on 10 July 2026, Circle received final OCC approval to establish First National Digital Currency Bank, N.A., a charter its own announcement says is "designed to enable future capabilities, including management of the USDC Reserve, which would bring those operations under federal regulatory oversight." [16] The reserve backing the second-largest dollar token on earth is being walked, deliberately, into the federal perimeter.

This report makes three arguments and then tries to break them. First: the statutory clock has not started, and the reporting forms the OCC has already printed are forms for issuers who do not exist. Second: the float is flat, the reserves are not what the headlines say they are, and the Treasury-demand story — on Treasury's own advisory committee's numbers — is a rounding error for the sovereign. Third: the GENIUS Act defines a permitted issuer as "a person formed in the United States," which makes a Cayman company categorically incapable of being one, and Cayman's own regulator's statistics show the virtual-asset register shrinking. [1] [19]

There is one date that could break all of this, and we will hand it over honestly at the end: roughly 20 September 2026, the last point at which a final rule could still pull the effective date forward.

The Clock That Never Started

Start with what the statute actually does. The GENIUS Act is Public Law 119-27, approved July 18, 2025. [1] Section 3(a), codified at 12 U.S.C. 5902, is the operative prohibition: "It shall be unlawful for any person other than a permitted payment stablecoin issuer to issue a payment stablecoin in the United States." [1] Section 2(23) then defines the term with a phrase that does more work than any other four words in the Act: a permitted payment stablecoin issuer means "a person formed in the United States" that is a subsidiary of an insured depository institution approved under section 5, a Federal qualified payment stablecoin issuer, or a State qualified payment stablecoin issuer. [1]

And Section 20 says when all of that switches on. The Act "shall take effect on the earlier of— (1) the date that is 18 months after the date of enactment of this Act; or (2) the date that is 120 days after the date on which the primary Federal payment stablecoin regulators issue any final regulations implementing this Act." [1] OCC Bulletin 2026-3 restates the identical test. [3]

Read prong (2) slowly. The trigger is final regulations. Now look at the record.

The OCC's rule is a Notice of Proposed Rulemaking — Docket ID OCC-2025-0372, RIN 1557-AF41, proposing amendments to 12 CFR Parts 3, 6, 8, 15 and 19, published 2 March 2026. [2] Its comment window is not merely open, it is not yet even dated: the DATES field reads "Comments must be received by [INSERT DATE 60 DAYS AFTER DATE OF PUBLICATION IN THE FEDERAL REGISTER]," and the OCC confirms it "issued a proposed rule on March 2, 2026, and is requesting public comment within 60 days of Federal Register publication." [2] [4] On 8 April 2026, Treasury announced that FinCEN and OFAC "issued a joint proposed rule" implementing the Act's anti-money-laundering and sanctions-compliance provisions. [10]

Proposal. Proposal. There is no final regulation from any primary Federal payment stablecoin regulator in the record. The 120-day clock has never started. Which means the operative effective date is the other prong — the 18-month backstop, which lands on 18 January 2027.

The consequence is unglamorous and total: no GENIUS requirement currently binds any issuer. Not the 1-to-1 reserve rule. Not the ban on paying yield. Not the reporting obligations. Not the prohibition in Section 3(a) itself. Every dollar-pegged token in circulation on 14 July 2026 was issued under arrangements that predate the statute, because the statute has not taken effect and nobody has been licensed under it.

1. 18 Jul 2025

   GENIUS Act signed — Public Law 119-27[1][3]

   Sec. 3(a): "It shall be unlawful for any person other than a permitted payment stablecoin issuer to issue a payment stablecoin in the United States." Every implementation clock in the Act runs from this date.

2. 12 Feb 2026

   OCC grants Bridge National Trust Bank (Stripe) preliminary conditional approval[5]

   Corporate Decision #1365. Final authorisation under 12 USC 27(a) awaits preopening requirements. The wrapper is being chartered onshore.

3. 25 Feb 2026

   OCC Bulletin 2026-3 announces the rulemaking[3]

   Restates the Sec. 20 test: effective on the earlier of 18 months after enactment, or 120 days after final regulations.

4. 2 Mar 2026

   OCC issues a NOTICE OF PROPOSED RULEMAKING — not a final rule[2][4]

   Docket ID OCC-2025-0372, RIN 1557-AF41. DATES field: "Comments must be received by [INSERT DATE 60 DAYS AFTER DATE OF PUBLICATION IN THE FEDERAL REGISTER]."

5. 8 Apr 2026

   FinCEN and OFAC issue a JOINT PROPOSED RULE[10]

   BSA and sanctions-compliance obligations for permitted payment stablecoin issuers — proposed, open for comment.

6. 1 May 2026

   OCC Bulletin 2026-24 prints the reporting forms[4]

   Form PS-01 (weekly confidential reserve reporting) and Form PS-02 (quarterly condition and income) — forms for a category of institution with zero members.

7. 10 Jul 2026

   Circle receives FINAL OCC approval for First National Digital Currency Bank, N.A.[16]

   A charter "designed to enable future capabilities, including management of the USDC Reserve, which would bring those operations under federal regulatory oversight."

8. 14 Jul 2026

   As-of date: no final rule, no permitted issuer[2][10][1]

   The 120-day trigger has not started. Count of approved permitted payment stablecoin issuers: zero.

9. ~20 Sep 2026

   Last date a final rule could still pull the effective date forward[1]

   Journal arithmetic: 18 January 2027 minus 120 days. Finals published after this cannot beat the backstop. This is the live falsifier.

10. 18 Jan 2027

    Sec. 20 backstop — the Act takes effect[1][3]

    Journal arithmetic: enactment + 18 months. This is the operative date unless final regulations arrive first.

11. 18 Jul 2028

    Delisting cliff — enactment + 3 years[1]

    Sec. 3(b)(1): US digital asset service providers may not offer or sell a payment stablecoin not issued by a permitted payment stablecoin issuer. The shot clock on any offshore wrapper.

The forms exist. The issuers do not.

The most quietly telling document in the whole file is OCC Bulletin 2026-24, issued 1 May 2026. It sets out the reporting architecture for permitted payment stablecoin issuers: each must file "a weekly confidential reporting form to the OCC for each payment stablecoin it issues" — Form PS-01, "Payment Stablecoin Activity and Reserve Weekly Reporting" — plus "a quarterly reporting form to the OCC," Form PS-02. Foreign payment stablecoin issuers registered with the OCC are subject to the same requirements. [4]

The Comptroller has printed weekly reserve-reporting forms for a category of institution that, as of this writing, has exactly zero members. That is not incompetence; it is a regulator building the plumbing ahead of the water. But it is also the cleanest available picture of where this actually stands. The scaffolding is up. The building is empty.

Two further dates fall out of the statute's own arithmetic. Section 3(b)(1) provides that "beginning on the date that is 3 years after the date of enactment of this Act, it shall be unlawful for a digital asset service provider to offer or sell a payment stablecoin to a person in the United States, unless the payment stablecoin is issued by a permitted payment stablecoin issuer." [1] Three years from 18 July 2025 is 18 July 2028 — a delisting cliff for any coin whose issuer is neither permitted nor compliant with the Act's foreign-issuer exception. And 120 days back from the 18 January 2027 backstop gives roughly 20 September 2026: the last date on which a final rule could still arrive early enough to matter.

The most senior American official statement in the record, meanwhile, is rhetoric rather than schedule. Secretary of the Treasury Scott Bessent, announcing the FinCEN/OFAC proposal: "This proposal will protect the U.S. financial system from national security threats without hindering American companies' ability to forge ahead in the payment stablecoin ecosystem." [10] No figure, no horizon, no methodology. When you strip the adjectives out of the official record, what remains is a proposal and a comment window.

The Float That Stopped Growing

Here is where the received narrative and the primary sources part company hardest.

The Federal Reserve has now published three separate readings of stablecoin market size inside four months, from three different desks, and they say the same thing. The Richmond Fed's March 2026 Economic Brief cites market capitalisation "now exceeding $300 billion" (as noted on CoinMarketCap as of March 3). [13] The Board's FEDS Note of 8 April 2026 puts the market at $317 billion as of April 6, 2026 — acknowledging "more than 50% growth since early 2025," but immediately qualifying it: growth "flatten[ed] out during the final quarter of 2025 and first quarter of 2026." [11] The May 2026 Financial Stability Report is blunter still. Stablecoin assets "have grown 16 percent from July 2025 to the end of last year. However, this rapid growth has moderated in recent months amid a broader decline in the price of crypto-assets—a pattern consistent with the fact that stablecoins are mostly used to facilitate crypto-trading activities—and the level is currently $320 billion, concentrated among the two largest issuers." [12]

Read that middle clause again, because it is the Fed doing something more aggressive than reporting a number. "Stablecoins are mostly used to facilitate crypto-trading activities" is the central bank rejecting the payments-adoption frame outright. On the Fed's account, the float went down because crypto prices went down — which is what you would expect of collateral for a trading book, and not what you would expect of a payments rail on its way to becoming plumbing.

Source: Richmond Fed Economic Brief 26-10 (Mar 2026); Board of Governors FEDS Note (8 Apr 2026); Federal Reserve Financial Stability Report (May 2026)

The issuers corroborate it. Tether's Q1 2026 report — the one with the record buffer and the billion-dollar quarter — describes its own token supply in language no growth company would choose: "USD₮ in circulation remained broadly stable at scale, with total token-related liabilities of approximately $183 billion as of March 31, 2026." [14] Broadly stable. That is the largest stablecoin on earth, in its own attestation, declining to claim growth.

The counter-datum, handled honestly

There is one number that appears to cut the other way, and it is Circle's. In Q1 2026, "USDC in circulation grew 28% to $77.0 billion; USDC onchain transaction volume grew 263% to $21.5 trillion." [6] [15] Both figures are real, audited-adjacent, and filed with the SEC. Both are also, on inspection, weaker than they look for this purpose.

The 28% is a year-over-year comparison — Q1 2026 against Q1 2025. It tells you what happened across twelve months, most of which sit inside the boom the Fed says has since ended. It tells you nothing whatsoever about the current run-rate. The $21.5 trillion is turnover, not float: velocity through the token, not the size of the token. And Circle's third headline — "USDC representing 63% of stablecoin transaction volumes in the first quarter" — is explicitly a share of transaction volume, not a share of the money. [15] On float, USDC's $77.0 billion sits against Tether's roughly $183 billion. [14]

Add the two issuers together and you get about $260bn — roughly 82% of the market the Fed measured on 6 April 2026, which corroborates the FSR's own phrase, "concentrated among the two largest issuers." [12] The residual, spread across every other stablecoin in existence, is roughly $57bn.

And there is one more thing the Fed does with stablecoins that deserves more attention than it gets: it counts them. Formally. They appear in the FSR among the "unlisted components of runnable money-like liabilities," in the company of "offshore money market funds, short-term investment funds, local government investment pools, and stablecoins." [12] The FEDS Note frames the concern as transmission, not containment: run risk can "rapidly propagate through interconnections both within digital asset ecosystem and across the traditional financial system." [11] The official sector's operative category for this asset class is not "payments innovation." It is "runnable."

What Is Actually In The Reserve

Section 4(a) of the Act is short and it is a fence. A permitted payment stablecoin issuer must "maintain identifiable reserves backing the outstanding payment stablecoins of the permitted payment stablecoin issuer on an at least 1 to 1 basis," and the permitted reserve assets are enumerated: US coin and currency or money at a Federal Reserve Bank; demand deposits or insured shares at an insured depository institution; "Treasury bills, notes, or bonds— (I) with a remaining maturity of 93 days or less; or (II) issued with a maturity of 93 days or less"; overnight repurchase agreements backed by Treasury bills of 93 days or less; and overnight reverse repos collateralised by Treasuries. [1]

Cash, insured deposits, very short Treasuries, and overnight Treasury repo. That is the whole list. Now measure the duopoly against it.

USDC: already GENIUS-shaped — and mostly repo

Circle's April 2026 reserve report, examined by Deloitte & Touche, discloses the reserve at two dates. USDC in Circulation was 77,745,231,766 at April 6, 2026 and 77,047,590,794 at April 30, 2026, against Fair Value of Assets Held in USDC Reserve of $77,821,343,057 and $77,123,668,726 respectively. [7] Deloitte's opinion is narrow and worth stating precisely — it is not a financial-statement audit but an examination of a single assertion: that "the Fair Value of Assets Held in USDC Reserve is equal to or greater than USDC in Circulation as of the Report Dates." [7]

Every Treasury CUSIP in the 30 April reserve matures between 05/05/26 and 07/23/26. The longest-dated bill in the book, 912797TP2, matures 84 days after the report date — comfortably inside the statute's 93-day cap. USDC's reserve is GENIUS-shaped before the rules bind. [7] [1]

But the composition is the finding, and it is not the one the market talks about. At 30 April 2026 the BlackRock-managed Circle Reserve Fund — a government money market fund subject to Rule 2a-7 under the Investment Company Act of 1940 [6] — held TOTAL U.S. TREASURY SECURITIES of 19,110,672,894 against U.S. Treasury Repurchase Agreements of 46,998,000,000, inside total fund assets of 66,367,076,329, alongside 10,567,044,499 of cash at regulated financial institutions. [7]

Against the total reserve, that is 60.9% overnight Treasury repo, 24.8% outright Treasury securities, and 13.7% bank cash. Circle held roughly two and a half times as much overnight repo as outright bills. The stablecoin that everyone describes as a great buyer of Treasury bills holds a quarter of its reserve in Treasury bills.

Worse for the demand story: the mix moves. Four weeks earlier, on 6 April 2026, the same fund held 25,643,128,238 of Treasury securities against 40,579,500,000 of repurchase agreements. [7] In the space of a single month, Circle rotated roughly $6.5bn out of outright bills and roughly $6.4bn into repo. Whatever that is, it is not a stable structural bid for government paper. It is a cash manager optimising, and it swings.

USDT: 1.04x on the books, 0.74x on quality

Tether's BDO attestation as of 31 March 2026 asserts total assets of US$191,767,741,495 against total liabilities of US$183,535,531,717, of which US$183,438,487,810 relate to digital tokens issued — assets exceeding liabilities by US$8,232,209,778, a record buffer. [14] On its face: over-collateralised, and profitable, with roughly $1.04 billion of net profit in the quarter. [14]

The Federal Reserve looks at the same balance sheet and reaches a different number. USDT, it finds, maintains "approximately 1.04x in reserves for each coin in circulation," but only "about 0.74x in assets qualifying as higher-quality reserves" — meaning Treasuries, repos backed by Treasuries, and bank deposits. USDC, by contrast, "maintains full 1.0x backing with higher-quality reserves." [11]

Both numbers are true, and the gap between them is itemised in Tether's own disclosure. "Precious metal holdings, consisting entirely of physical gold, amounted to approximately $20 billion. Bitcoin holdings stood at approximately $7 billion." [14] Neither gold nor bitcoin appears anywhere on the Section 4(a) list. [1] Tether is over-collateralised in accounting terms and under-collateralised in GENIUS terms, and the difference is roughly $27bn of assets the statute does not recognise as backing.

A Rounding Error For The Sovereign

The macro claim attached to stablecoins — that they are becoming a structurally important buyer of US government debt — has one authoritative official-sector measurement in the record, and it comes from inside the Treasury building.

The Treasury Borrowing Advisory Committee's Q1 2026 Charge 2 presentation, on trends in demand for US Treasury securities, contains a slide on stablecoins. It says: "Stablecoin issuers currently hold <1% of Treasuries outstanding." [8] For the two majors it reports that "T-bills constitute 53% of assets," notes a "$70B increase in T-bill holdings since 2022," and charts asset composition from 2022 to 2025 at a CAGR of 35% — on an axis that tops out at $300bn, with 2025 data as of 9/30/2025. [8]

That is the anchor. Any macro claim about stablecoins and the Treasury market that does not start from "under 1% of Treasuries outstanding" is starting from somewhere other than the evidence.

Two further features of the TBAC record discipline the framing further. The first is that TBAC's demand claim is conditional, and the condition is doing enormous work: "Stablecoins have the potential to increase demand for short-end Treasury issuance should stablecoin growth be driven by new offshore demand currently not in USD." [8] If instead the growth comes out of bank deposits and money market funds — dollars that were already buying bills through a different wrapper — then it is substitution, and the sovereign gets nothing new. TBAC's own summary ranks the channel accordingly: "Shifts in the Fed SOMA portfolio composition represents a key structural shift increasing demand for T-bills. Money markets are likely to continue to be a source of demand for T-bills, and stablecoins are an emerging demand area." [8] Emerging. Third on a list of three.

The second is a negative finding, and it is the sharpest instrument in this section. The published Minutes of the TBAC meeting of February 3, 2026 do not discuss stablecoins at all. [9] The analysis lives in a presenting members' deck; it does not appear in the Committee's recommendations to the Secretary. What the Committee did tell the Secretary is that current coupon auction sizes leave Treasury well positioned through FY2026, "with changes in Treasury bill supply likely being adequate to address any changes in financing needs." [9] Stablecoin demand is not, on this record, a financing variable. It is a slide.

As Tether discloses itHalf the exposure is indirectAll of it is indirectReset

Tether disclosed T-bill exposure ("direct and indirect")fact141$bn

Share of that exposure assumed to be OUTRIGHT billsassumption100%

USDC total reserve assets (30 April 2026)fact77.1$bn

USDC reserve held in outright Treasury securitiesfact24.8%

Total stablecoin float (Fed, May 2026)fact320$bn

Tether: implied outright bills$141.0$bn

Circle: outright Treasury securities (Deloitte-examined)$19.1$bn

Combined outright bill holdings of the duopoly$160.1$bn

Tether's share of the entire bill story88.1%

Outright bills held per $1 of stablecoin float0.50×

The only outright-versus-repo split examined by an accounting firm is Circle's: 19,110,672,894 of Treasury securities against 46,998,000,000 of overnight Treasury repurchase agreements at 30 April 2026 (24.8% and 60.9% of the reserve respectively). Tether's $141bn is explicitly "direct and indirect exposure," and TBAC's "T-bills constitute 53% of assets" blends both issuers. Whatever the slider setting, the official-sector anchor does not move: TBAC's Q1 2026 Charge 2 presentation states that "Stablecoin issuers currently hold <1% of Treasuries outstanding."

The definitional trap

There is a further problem, and it is the reason so many published stablecoin-Treasury numbers are larger than they should be: the phrase "holds T-bills" means different things in different documents, and almost nobody says which one they mean.

TBAC's 53% is a blended figure across Tether and Circle. [8] Tether's headline is explicit that it is a broad measure: "direct and indirect exposure to U.S. Treasury bills amounted to approximately $141 billion" as of 31 March 2026 — direct and indirect. [14] (Tether also characterises itself as "the 17th largest holder of U.S. Treasuries globally" — its own claim, not an official ranking.) Circle, uniquely, has line items examined by an accounting firm that separate outright bills from repo, and there the outright figure is 19,110,672,894 against 46,998,000,000 of repurchase agreements. [7]

So: the "stablecoins are a giant T-bill buyer" narrative is, on the documented numbers, mostly a Tether narrative — and even at Tether it is an indirect figure. Every sentence in this report about stablecoin bill demand states which measure it uses, because the two differ by a factor that matters.

None of which is to say the mechanism is fictional. The Richmond Fed's March 2026 brief lays it out cleanly, and its theoretical conclusion is genuinely favourable to the dollar: "Reserve-backed stablecoins increase demand for U.S. Treasuries, while crypto-backed ones reduce it," with reserve-backed issuance becoming dominant as adoption broadens and "putting downward pressure on the natural rate of interest." Its closing line is worth quoting against ourselves: "What initially appears to be a challenge to the dollar can — under plausible institutional arrangements, such as those required by the GENIUS Act — become a force that strengthens it." [13]

That is a model, and it may well be right about the destination. It is not a measurement of where we are. Where we are is under 1% of Treasuries outstanding, a flat float, a reserve that is majority overnight repo at one issuer and a quarter non-permitted assets at the other — and a statute that caps permitted Treasury reserves at 93 days or less, which means that however large this channel grows, it can never become demand for duration. [1] It is a bill-and-repo bid by legislative design.

The Wrappers Are Onshoring, And Cayman Is Foreclosed

Which brings us to the second half of the received story: that the Cayman Islands are emerging as a significant domicile for the legal wrappers of dollar issuance. This is not a claim that is weakly supported. It is a claim foreclosed by the text of the statute and contradicted by every observable flow.

The law

Section 2(23): a permitted payment stablecoin issuer is "a person formed in the United States." [1] A Cayman-formed company is categorically incapable of being one. Not disadvantaged — incapable. There is no waiver, no election, no subsidiary trick that changes where the issuing person was formed.

The only door for a non-US issuer is Section 18, and it has four locks. The foreign issuer must be regulated by a foreign regime that the Secretary of the Treasury determines "is comparable to the regulatory and supervisory regime established under this Act, including, in particular, the requirements under section 4(a)"; it must be "registered with the Comptroller"; it must hold "reserves in a United States financial institution sufficient to meet liquidity demands of United States customers"; and its home country must be neither comprehensively sanctioned nor a jurisdiction of primary money laundering concern. Treasury may make the comparability determination "only upon a recommendation from each other member of the Stablecoin Certification Review Committee," with a Federal Register justification published first. [1]

Note what Section 18 names "in particular": section 4(a). Reserves. Now consider what Cayman has to offer on that dimension. Appleby's 2026 guide to blockchain and crypto-assets in the jurisdiction states it twice: "The Cayman Islands does not have a standalone stablecoin regulatory framework of the kind contemplated by the EU's Markets in Crypto-Assets Regulation (MiCA)," and "The Cayman Islands does not impose mandatory reserve or backing asset requirements on stablecoin issuers as a condition of VASP Act registration." CIMA has issued no stablecoin-specific guidance. [25]

On the single dimension the GENIUS Act legislates hardest, Cayman has nothing to compare. A comparability determination is not merely unlikely; there is no counterpart regime for Treasury to assess. CIMA's own VASP landing page confirms the absence from the other direction: it is a navigation hub, with links to registration requirements and statistics, no stablecoin category, no issuer licensing tier, no issuer list, and an enquiries mailbox at the "VASP & Fintech Innovation Unit." [31] There is no supervisory product line to point at.

The flows

Now the numbers, from the regulator itself. CIMA's published VASP Registration Statistics show registered entities at 18 in Q4 2025 — against 19 in Q3, 19 in Q2, and 20 in Q1 2025. [19] Across the year in which the GENIUS Act was signed, Cayman's virtual-asset register went down by two.

The service-line detail is worse. Virtual Asset Custody Services fell to 7 in Q4 2025 from 13 in Q1 2025. Virtual Currency to Virtual Currency/Fiat Conversion fell to 10 from 13. "Participating in financial services relating to virtual assets" fell to 4 from 7. Virtual Asset Trading Platforms number 3. [19] And the category that would capture stablecoin issuance — "Issuance of virtual assets" — has exactly 1 registrant, and has had exactly 1 in every quarter of 2025 and in 2024. [19] One registrant is the entire documented Cayman virtual-asset issuance industry.

Nor is the regime that supervises them mature. The VASP Act phased in slowly: Phase 1, registration with an AML/CFT focus, came into effect on 31 October 2020; Phase 2, which "introduced a licensing regime for virtual asset trading platforms and virtual asset custodians," commenced on 1 April 2025 — three and a half months before GENIUS was signed. [26] And CIMA's own thematic desk-based review of 11 regulated VASPs, conducted September 2024 to February 2025 with findings published in November 2025, reads as a list of gaps: 27% of VASPs reviewed did not meet the three-director requirement; 36% had no formal succession planning; 27% had not appointed a qualified CISO or CIO; 40% had inadequate virtual asset custody policies; and "a staggering 82% of VASPs reviewed lacked any cybersecurity insurance." [26] That is the supervisory reality behind the phrase "institutional-grade offshore crypto hub."

The vector runs the other way

Meanwhile, the actual issuers are doing the opposite of what the hypothesis predicts.

Circle took final OCC approval on 10 July 2026 for First National Digital Currency Bank, N.A., operating as Circle National Trust — a charter explicitly "designed to enable future capabilities, including management of the USDC Reserve, which would bring those operations under federal regulatory oversight." [16] The USDC reserve already sits in an American money market fund: the BlackRock-managed Circle Reserve Fund, subject to Rule 2a-7, holding $66.5 billion at 31 March 2026. [6] USDC itself is issued by "Circle Internet Financial, LLC and Circle Internet Financial Europe SAS" — a US limited liability company and a French société par actions simplifiée. [7] And the word "Cayman" appears zero times in Circle's Form 10-Q for the quarter ended 31 March 2026. [6]

Stripe is doing the same thing. On 12 February 2026 the OCC granted preliminary conditional approval to Bridge Ventures LLC — a Stripe, Inc. company — to establish Bridge National Trust Bank, New York (Corporate Decision #1365, Control Nos. 2025-Charter-343543 and 2025-Waiver-343824), with final authorisation to commence business awaiting preopening requirements. [5]

And when the largest offshore issuer in the world went looking for a home jurisdiction, it did not pick George Town. USD₮ is issued by "Tether International, S.A. de C.V." — an El Salvador entity. [14]

The OCC's own proposed scope makes the direction of travel explicit: proposed part 15 would reach national banks and their subsidiaries, Federal savings associations, Federal branches, "foreign payment stablecoin issuers," nonbank entities seeking Federal qualified payment stablecoin issuer status, and certain State qualified issuers. [2] The regulator that would supervise a foreign issuer selling into the United States is the Comptroller of the Currency. Not CIMA.

	United States (OCC / GENIUS Act)	Cayman Islands (CIMA / VASP Act)
Can a company formed here be a permitted payment stablecoin issuer?[1]	Yes — Sec. 2(23) requires "a person formed in the United States"	No — categorically incapable. Sec. 2(23) forecloses it; there is no waiver or election
Standalone stablecoin regime[1][25][31]	GENIUS Act, P.L. 119-27 — effective 18 Jan 2027 at the latest	None. "The Cayman Islands does not have a standalone stablecoin regulatory framework" (Appleby). CIMA has issued no stablecoin-specific guidance
Mandatory reserve / backing requirements[1][25]	Sec. 4(a): 1-to-1, limited to cash, insured deposits, Treasuries of ≤93 days, and overnight Treasury repo	"Does not impose mandatory reserve or backing asset requirements on stablecoin issuers as a condition of VASP Act registration"
Registered issuers in the relevant category[2][10][19]	0 permitted payment stablecoin issuers — no final rule has issued	1 registrant under "Issuance of virtual assets" — and exactly 1 in every quarter of 2025 and in 2024
Direction of the register[16][5][19]	Circle: final charter, 10 Jul 2026. Bridge/Stripe: preliminary charter, 12 Feb 2026	VASP register: 20 (Q1 2025) → 18 (Q4 2025). Custody: 13 → 7. Conversion: 13 → 10
Where the two majors' wrappers actually sit[6][7][14]	Circle Internet Financial, LLC (US) and Circle Internet Financial Europe SAS; reserve in the BlackRock-managed, Rule 2a-7 Circle Reserve Fund ($66.5bn)	Zero occurrences of the word "Cayman" in Circle's Form 10-Q. Tether issues from El Salvador — Tether International, S.A. de C.V.
Who supervises a foreign issuer selling into the US[2][1]	The OCC. Proposed 12 CFR part 15 expressly reaches "foreign payment stablecoin issuers"	Not CIMA. Sec. 18 requires OCC registration plus a Treasury comparability determination reaching "in particular, the requirements under section 4(a)"
Supervisory track record in virtual assets[4][26]	Weekly Form PS-01 reserve reporting and quarterly Form PS-02 already drafted	CIMA thematic review of 11 VASPs: 27% missed the three-director rule, 40% had inadequate custody policies, "a staggering 82% ... lacked any cybersecurity insurance"

Far from making Cayman an emerging domicile, the GENIUS Act puts an offshore wrapper on a shot clock. From 18 July 2028, US digital asset service providers may not offer or sell a payment stablecoin that is not issued by a permitted payment stablecoin issuer. [1] An offshore issuer has two years to become a US-formed permitted issuer, or to get its home regime certified as comparable by a Treasury Secretary acting on the unanimous recommendation of the Stablecoin Certification Review Committee — in a jurisdiction that today has no reserve rules at all.

The Kernel: Cayman Is Writing Fund Law, Not Money Law

There is one part of the Cayman story that is real, and it deserves to be stated at full strength before it is measured.

In March 2026, Cayman brought into force three coordinated amendments — to the Mutual Funds Act, the Private Funds Act, and the Virtual Asset (Service Providers) Act — creating a statutory framework for tokenised fund structures. [22] [23] Appleby dates entry into force to 24 March 2026 and calls it "the most significant product-level development in Cayman funds regulation since the Private Funds Act came into force in 2020." [25] That is not marketing; it is a serious law firm's serious assessment, and the mechanism is genuinely elegant.

The core move is a carve-out, not a new licence. The VASP Act was amended so that "tokenised private funds and mutual funds registered with CIMA are excluded from Virtual Asset (Service Providers) Act," removing the dual-licensing risk that "had previously slowed institutional decisions." [29] [21] A "digital equity token" is defined as "a digital representation of the whole of an equity interest held by an investor in a mutual fund" — the underlying share, not the token, remains the source of legal rights, and tokens may transfer only "with operator approval, in accordance with the fund's offering and constitutional documents." [29] [27] CIMA gains express supervisory powers, including "the authority to carry out inspections of the underlying technology and digital token transactions." [23]

Note what that is: a wrapper for fund shares, transferable only with the operator's consent. It is the structural opposite of a bearer-instrument stablecoin.

And there is a reason this business exists at all, and it sits in the GENIUS Act itself. Section 4(a)(11): "No permitted payment stablecoin issuer or foreign payment stablecoin issuer shall pay the holder of any payment stablecoin any form of interest or yield (whether in cash, tokens, or other consideration) solely in connection with the holding, use, or retention of such payment stablecoin." [1] Yield is legislated out of the coin. So it moves next door, into the tokenised money-market fund. Cayman is not building a rival to the dollar wrapper; it is building the yield-bearing thing that sits beside it because the statute forbade the coin from paying.

Now measure it

Twelve funds. "The number of tokenised funds registered with CIMA has risen to 12 since Cayman's statutory framework for tokenised fund structures came into force in March 2026." [21] [24] Against a domicile that held 31,145 regulated funds at the end of Q2 2026 — 18,132 private funds and 13,013 mutual funds. [20] [21] Twelve out of 31,145 is 0.04% of the fund population.

Three qualifications compound that. First, the count of twelve comes from Cayman Finance, the jurisdiction's promotional body — not from CIMA, whose published investment statistics carry no tokenised-fund line item at all, and not from Walkers, whose note on the framework does not specify how many tokenised funds exist or quantify market demand. [20] [29] Second, the registrations are provisional. CIMA has not finalised its permanent rule; in April 2026 it circulated "a preliminary questionnaire and a limited set of revised conditions" that serve "as the basis for approval of tokenised fund registration applications" until the new rule finalises. [28] The regime registering these funds is itself still in temporary measures.

Third — and this is the one that would embarrass anybody who built a crypto story on the big number — Cayman Finance itself does not attribute its growth to tokenisation. "Cayman added 547 new funds in the first half of the year, comfortably ahead of the 448 recorded in the whole of 2025. The growth comes from institutional investors increasing their private market allocations, greater demand for continuation vehicles and a steady flow of new hedge fund launches." [21] Tokenisation is named as "the newest fund trend," not as the engine. Private fund registrations rose by 410 in six months, and 56% of emerging hedge fund managers domicile their flagship fund in Cayman. [21] That is a private-credit and hedge-fund story. Putting a crypto frame on the 31,145 number is the single most likely way this subject gets reported wrong.

The anatomy of the one honest datapoint

USYC is the clearest documented Cayman wrapper anywhere in the stablecoin complex, and on inspection it argues against the thesis it is usually cited to support.

Circle's product page: "Each USYC token serves as a digital representation of a share of the Hashnote International Short Duration Fund Ltd. (the 'Fund'), a Cayman Islands registered mutual fund." [17] But the token layer is not Cayman. Circle's developer documentation: the token is "issued by Circle International Bermuda Limited, regulated by the Bermuda Monetary Authority (BMA), and represents an interest in a Cayman Islands mutual fund licensed by the Cayman Islands Monetary Authority (CIMA)." [18] Bermuda issues; Cayman holds the fund. And it is closed to Americans by construction: "USYC is available only to entities that are not U.S. Persons, as defined in Regulation S under the Securities Act of 1933." [18]

Its AUM is $312,976,695.16. [17] Against USDC's $77.0 billion, that is 0.41%. [6] The Cayman-domiciled leg of Circle's tokenised-cash stack is roughly four tenths of one percent the size of its stablecoin — and it cannot be sold to a single American.

Even the flagship exhibit is contested. Cayman Finance says Fidelity domiciled its first tokenised fund in Cayman — "Fidelity's decision to domicile its first tokenised fund here shows the legislative reforms are working as intended" — and lists the Fidelity USD Digital Liquidity Fund among the twelve. [21] The independent trade coverage of the fund's Moody's AAA-mf rating, awarded 13 May 2026, does not state a Cayman domicile, and identifies the underlying LVNAV fund — approximately $7 billion in assets — as Irish. [30] We attribute the domicile. We do not assert it.

And the prize itself is smaller than the one it is narratively attached to. Global tokenised fund AUM "reached over $33 billion globally by July 2026," per Cayman News Service — against a stablecoin market the Fed puts at $317 billion in April and $320 billion in May. [24] [11] [12] Total victory in tokenised funds worldwide would be roughly a tenth of the stablecoin pool, and Cayman currently holds twelve of them. The forward figure you will see quoted back — that tokenised fund AUM "could hit $600 billion by 2030" — is a WealthBriefing projection, relayed by Cayman News Service. [24] It is an estimate produced by a trade publisher, not a measurement, and it should never be printed without that label.

One genuinely well-documented Cayman digital-asset franchise does exist, and it sits outside both stablecoins and funds: "The Cayman foundation company has become the dominant structural choice for DeFi protocol governance entities and DAO vehicles." [25] It is a governance wrapper. It has nothing to do with dollar issuance or Treasury reserves.

What Would Break This

The case against this report is not stupid, and it deserves its strongest form.

The infrastructure is unmistakably being built. The OCC has published the scope, the capital treatment, the custody rules and the reporting forms. [4] Circle's reserve is already compliant in shape — every bill inside the 93-day cap, the whole thing sitting in a Rule 2a-7 government money market fund. [7] The Richmond Fed's model says reserve-backed issuance becomes dominant in the long run and, under arrangements like those GENIUS requires, strengthens the dollar rather than threatening it. [13] TBAC calls stablecoins "an emerging demand area." [8] The tokenised-fund pool is over $33 billion and Cayman's framework for it is well-drafted and unanimously supported. [24] [29] Everything this report describes as "not yet" is, plausibly, "not yet."

And "not yet" has a specific mechanism by which it becomes "now": Section 20's 120-day trigger. [1] A final rule in the Federal Register, followed by the OCC's first approved permitted payment stablecoin issuer, would arm the statute early and make the count of licensed issuers non-zero. The spine of this report — nothing binds, nobody holds the licence — would then need rebuilding, not hedging. Watch for it before roughly 20 September 2026, the last date on which finals could still beat the 18 January 2027 backstop.

Four other tests would move us, and we state them without wriggle room. If aggregate stablecoin supply makes a decisive new high above roughly $330bn before Q4 2026, with Tether's digital-tokens-issued line rising again, then "stalled" is factually backwards. If Treasury publishes a Section 18 comparability determination naming the Cayman Islands — or if CIMA legislates reserve, redemption and attestation rules that could support one — the central finding flips. If any named stablecoin issuer redomiciles to Cayman and registers under Section 18, the onshoring thesis breaks. And if CIMA's tokenised-fund count clears 50 and appears as an official line item in the regulator's own statistics rather than a promotional body's press release, then "too small to carry the sentence" stops being true.

Sustained "no" on all of them, and what remains is what the primary sources actually say. A statute that has not taken effect. A register of permitted issuers with no entries. A float that two Federal Reserve publications a month apart describe as flat, and that its largest issuer calls "broadly stable." A Treasury channel that Treasury's own advisory committee sizes at under 1% of Treasuries outstanding, capped by law at 93 days, and mostly transacted through overnight repo rather than the bills everybody talks about. And an offshore jurisdiction whose virtual-asset register shrank, whose issuance category contains one company, and whose genuinely new digital-asset franchise is twelve provisional fund registrations sold to people who are not allowed to be American.

Cayman is writing a wrapper for tokenised money-market fund shares. It is not writing the wrapper for the dollar going on-chain. That wrapper is being chartered, right now, by the Comptroller of the Currency — and nobody holds it yet.