Executive summary
Three of the largest regulated markets now have stablecoin frameworks that are either live or written into law: the EU's MiCA, the US GENIUS Act, and the UK's FSMA 2000 (Cryptoassets) Regulations 2026. The supplied data lets me compare them on the dimensions that actually constrain a builder — licensing, reserves, redemption, reporting, capital, and conduct restrictions.
The headline is convergence on the basics and divergence on the edges. All three require full 1:1 backing, par-value redemption, and segregated reserves held away from the issuer. But they split on whether you can pay holders interest, how they treat algorithmic designs, and whether foreign-currency stablecoins get capped. In my view those edges, not the shared core, are where issuance decisions will be made.
What the data shows
Start with what is identical. Every framework demands at least 1:1 reserves in high-quality liquid assets, segregated from operating funds. MiCA requires EMTs to be backed 1:1 by funds with at least 30% in bank deposits (60% for significant EMTs) (EBA). The GENIUS Act requires reserves "on at least a 1:1 basis" in cash, insured deposits, short-term Treasuries, and qualifying repos, with rehypothecation prohibited except to meet redemptions (S.1582). The FCA's CP25/14 proposal mirrors this with on-demand deposits and sub-one-year government debt, held on statutory trust by unconnected custodians (FDIC FR notice).
The divergence starts with timing. MiCA's stablecoin titles have applied since 30 June 2024 and its CASP transitional period ends 1 July 2026 — it is the only one of the three that is fully operative (ESMA). The GENIUS Act was signed 18 July 2025 but takes effect no later than 18 January 2027, or 120 days after final rules; the OCC, FDIC, and FinCEN have issued proposals through early 2026 while the Federal Reserve has not yet proposed (OCC bulletin). The UK made its statutory instrument on 4 February 2026 but does not fully commence until 25 October 2027. So a builder in mid-2026 is operating against one live regime and two that are legally certain but mechanically unfinished.
Three philosophies of reserve composition
The reserve rules look similar but encode different anxieties. MiCA's deposit floor — 30% of an EMT reserve in bank deposits, 60% for significant ones — pushes balances back into the banking system, which I read as a financial-stability preference over yield optimization. The GENIUS Act leans the other way: it favors Treasuries and repos, and the OCC proposal layers on a $5M minimum capital floor plus twelve months of operating expenses in liquid assets during a three-year de novo period (Sullivan & Cromwell). The UK's statutory-trust structure is the most legally explicit about holder protection in insolvency.
MiCA also carries the most prescriptive own-funds rule: ART issuers hold the highest of €350,000, 2% of average reserve assets (3% if significant), or a quarter of fixed overheads, with authorities able to add up to 20% (A&O Shearman). The US equivalent is still being set by rule. The practical read: MiCA gives you a number today, while the GENIUS framework gives you a direction and asks you to wait for the rulemaking to finish.
The interest prohibition fault line
The sharpest legal divergence in the data is interest. MiCA prohibits issuers and CASPs from granting any interest or benefit linked to holding duration (Title III commentary). The GENIUS Act bars issuers from paying yield for merely holding, and the OCC proposal extends that presumption to indirect payments through affiliates (Oxford Business Law Blog). The UK's CP25/14 is still in proposal form on conduct. On paper this looks like alignment, but the US carve-out matters: the ban targets the issuer, not necessarily an exchange or DeFi protocol routing reward to the holder. The Oxford analysis frames this as a "regulatory maze," and I think that is the right word — the prohibition's perimeter, not its existence, is the open question, and it is where product teams will probe.
Algorithmic designs and the foreign-currency cap
Two design questions separate the regimes further. On algorithmic stablecoins, MiCA is categorical: purely algorithmic tokens cannot qualify as ARTs or EMTs and may not be marketed as stablecoins. The GENIUS Act, by contrast, commissions a study rather than imposing a statutory ban. That is a genuine philosophical fork — prohibition versus deferred judgment — and it leaves a category legal in one market and excluded in another.
The second is MiCA's foreign-currency cap, which has no counterpart in the supplied US or UK data. A non-euro EMT or ART used widely as a means of payment must halt new issuance and file a remediation plan once quarterly-average daily transactions used as a means of exchange exceed €200 million or 1 million transactions inside the EU (The Block). For dollar-denominated stablecoins — which dominate global volume — this is a hard ceiling on EU payment use that simply does not exist in the US framework. In my view this is the single most consequential asymmetry in the dataset for cross-border issuers.
What it means
For builders, the data points to picking a home regime and designing to its strictest binding constraint, then treating the others as overlays. If you are euro-native, MiCA is your operative rulebook today and the €200M/1M-transaction cap shapes any non-euro ambition. If you are dollar-native, the GENIUS Act gives you a federal/state choice — note the $10B ceiling that forces state-qualified issuers up to federal supervision, and the $50B line where monthly attestations become annual audited financials.
For institutions evaluating counterparties, the reporting thresholds are a useful sorting mechanism: MiCA's €100M quarterly-reporting trigger and the GENIUS Act's tiered attestation regime mean transparency scales with size by design. As a regulated venue, that is the kind of structure I would rather build on than around.
The forward question is sequencing. With MiCA live, the US effective by January 2027, and the UK commencing October 2027, the next eighteen months are a window where the same token can face three different statuses at once. I'd watch whether the US final rules narrow the indirect-interest perimeter and whether other jurisdictions copy MiCA's foreign-currency cap — because if they do, the dollar stablecoin's global default status becomes a jurisdiction-by-jurisdiction negotiation rather than a given.
Methodology
This report interprets structured regulatory data supplied for three frameworks where detailed, citable provisions were available: the EU's MiCA (Titles III and IV), the US GENIUS Act, and the UK's FSMA 2000 (Cryptoassets) Regulations 2026. Figures — reserve ratios, thresholds, caps, and dates — are drawn directly from that dataset and the linked primary and secondary sources. Where rules are proposed rather than final (notably the US implementing rulemakings and the FCA conduct rulebook), I have flagged them as such; those provisions may change before commencement. This is regulatory analysis, not legal or financial advice, and it does not assess any specific token. Comparisons across additional jurisdictions would require equivalent structured data, which was not part of this set.