On 1 July 2026, MiCA's stablecoin rules moved from phase-in to full enforcement across the EU and USDT quietly disappeared from the last major regulated venue still offering it. Revolut, which had held out longer than Coinbase, Binance, Kraken and the rest, announced it would halt USDT purchases from 6 July, stop accepting new deposits after 30 July, and auto-convert any balance left by 31 August into fiat at whatever rate applied on the day (TechTimes, July 2026).
Circle secured e-money institution authorisation from France's ACPR in July 2024, becoming the first global stablecoin issuer licensed under MiCA. That authorisation covers both USDC and EURC. Tether chose not to pursue it, citing MiCA's requirement that at least 60% of a significant EMT issuer's reserves sit in EU bank deposits, which conflicts with their Treasury-heavy model.
Once full MiCA enforcement began on 1 July 2026, licensed EU platforms had no legal basis to keep offering USDT to retail users; delisting was automatic.
In the US, the GENIUS Act (S. 1582), the federal stablecoin framework signed into law in July 2025, limits an issuer's permitted reserves to cash, insured deposits, short-dated Treasury bills, repos backed by Treasuries, and government money market funds. USDT doesn't meet that criteria: Tether is BVI-incorporated without a US banking charter, and its gold and Bitcoin holdings fall outside the reserve categories the Act allows.
If your business takes stablecoins from EU counterparties, your decision is already narrowed for you. But regulation is just one consideration - you also have to consider reverse backing, how rigorous attestation really is, and where and how coins move at scale.
What Actually Backs Each Coin
Both USDT and USDC are marketed as 1:1 dollar-backed, but the composition and the way each issuer proves it differs.
With USDC (Circle), the reserves sit in cash and short-dated US Treasuries, split roughly 80% Treasuries to 20% cash, held via the Circle Reserve Fund, a SEC-registered government money market fund managed by BlackRock. Circle publishes monthly attestations from Deloitte. USDC has had one depeg event in March 2023, tied to Silicon Valley Bank exposure and re-pegged within 72 hours. (We cover that episode in more detail in What Is a Stablecoin? Types, Backing and How They Stay Pegged.)
Tether's Q1 2026 attestation, published by BDO Italia, reports around $141 billion in US Treasury exposure against roughly $183.5 billion in circulating USDT (about 80–83% of reserves) with the remainder split across overnight repo, cash, roughly $8–10 billion in gold, and a similar amount in Bitcoin. Excess reserves stood at a record $8.23 billion for the quarter, and Tether posted $1.04 billion in net profit.
Just the Treasury holding on its own makes Tether one of the twenty largest holders of US government debt globally.
Remember: a BDO attestation is a point-in-time confirmation of a specific claim, not a full audit of controls across a period. Deloitte's monthly cadence and BDO's quarterly cadence are both attestations rather than audits in the traditional sense. Circle simply reports more often and holds a narrower, more liquid asset mix. Neither issuer has published a full annual audit covering their total reserves.
Regulatory Status: Where the Real Split Is
Circle secured e-money institution authorisation from France's ACPR in July 2024, becoming the first global stablecoin issuer licensed under MiCA. That authorisation covers both USDC and EURC. Tether chose not to pursue it, citing MiCA's requirement that at least 60% of a significant EMT issuer's reserves sit in EU bank deposits, which conflicts with their Treasury-heavy model.
Once full MiCA enforcement began on 1 July 2026, licensed EU platforms had no legal basis to keep offering USDT to retail users; delisting was automatic.
In the US, the GENIUS Act (the federal stablecoin framework passed in late 2025) has effectively codified standards that favour Circle's existing structure. USDT doesn’t meet GENIUS Act criteria: Tether is BVI-incorporated without a US banking charter, and its gold and Bitcoin holdings fall outside what the Act treats as qualifying reserve assets.
If a counterparty is EU-regulated, USDC/EURC is now close to the only compliant option. Outside the EU, USDT still dominates by volume; it's the larger of the two coins by circulating supply ($183.5 billion vs USDC’s ~$78 billion as of Q1/Q2 2026) and remains the default settlement asset across most non-EU exchanges, OTC desks and payment-agent flows.
Where Each Coin Actually Moves at Scale
Regulatory status aside, the two coins behave differently once you look at where volume actually sits. USDT's dominance is concentrated on Tron, where transfer fees average a few tenths of a cent and retail-sized transfers are common across Latin American, African and Southeast Asian corridors. USDC skews toward Ethereum and its rollups, plus native issuance across roughly two dozen chains connected by Circle's Cross-Chain Transfer Protocol.
A 2026 Harvard Business School working paper on cross-border payment rails found that gas fees on the underlying networks are, in practice, close to irrelevant to the total cost of moving stablecoins. Average per-transaction fees ranged from about $0.0003 on some Ethereum layer-2 networks up to roughly $0.53 on Tron, but even the higher end is trivial against a typical business-sized transfer.
Even on relatively expensive Layer 1 networks such as Ethereum, average transaction fees are well below one dollar, while fees on Layer 2 networks and alternative chains are typically only a few cents or less. The same research is blunt about where the real cost sits instead: centralised crypto exchanges effectively play a role analogous to correspondent banks, creating additional "hops" for fiat on- and off-ramping and reintroducing layers of intermediation into what is often described as a disintermediated payment rail.
That matches what we actually see processing this flow day to day. On the OTC side of our business, off-ramping (a client sends us USDC or USDT and we settle euro to their bank account) makes up the large majority of volume; on-ramping the other direction is a smaller slice.
Average ticket size across that flow runs in the low six figures per transaction. At that size, gas fees genuinely don't register. What does register is liquidity: we run trades against a credit line, and the practical constraint isn't which coin arrived, it's whether we've posted enough margin to cover the exchange side of a large ticket without the client waiting.
Trade size relative to available credit capacity matters: thin liquidity can mean delays, and manual execution steps add friction on top of that, regardless of coin choice.
The practical read for a business choosing between the two: coin choice is rarely the deciding factor on cost or speed once you're moving business-sized amounts. What actually decides it is (a) whether your counterparty or destination market is EU-regulated, which increasingly forces USDC/EURC, and (b) whether your settlement partner has the liquidity depth to clear your ticket size without you waiting on it.
Volatility and Peg Stability
Both coins are among the most stable stablecoins in circulation, but neither is flawless. USDC's SVB-related depeg in March 2023 was the more severe of the two recorded incidents, dropping to roughly $0.87 before recovering within three days once Circle's exposure was clarified. USDT wobbled to around $0.95 in May 2022 during a broader market stress event and recovered within hours.
Ten-plus years of continuous operation for both coins, across multiple market crashes and exchange failures, is the stronger signal than either isolated incident, but it's worth knowing both have been tested and both recovered without a sustained break from parity.
Which to Accept: A Decision Guide
| Factor | Lean USDC | Lean USDT |
|---|---|---|
| EU counterparties or EU-regulated platforms | Yes — sole MiCA EMT-authorised dollar stablecoin as of July 2026 | Not usable on licensed EU venues post-enforcement |
| US-regulated counterparties (GENIUS Act scope) | Yes — structure aligns with GENIUS Act standards | No — BVI structure and reserve mix fall outside scope |
| High-volume retail/emerging-market corridors (Latin America, Africa, SE Asia) | Workable, but thinner liquidity | Deeper liquidity, especially on Tron |
| Reserve transparency preference | Monthly Deloitte attestation, narrower asset mix (cash + Treasuries only) | Quarterly BDO attestation, broader mix including gold and Bitcoin |
| Settlement speed/gas cost on-chain | Negligible either way — network fees are not the deciding factor | Negligible either way — network fees are not the deciding factor |
| Counterparty already holds the coin | Match it — converting between the two adds an unnecessary FX/exchange-fee leg | Match it — same logic applies |
The short version: if your flow touches the EU, the regulatory answer is largely made for you. Outside the EU, the choice comes down to where your counterparties already sit and which exchange rails you're using to convert, not the coin's technical merits, which are close enough for either to work.
Related reading: How Pay-by-Crypto Works for B2B: Settlement, Chargebacks and Reconciliation · How to Accept Crypto Payments Without Holding Crypto (Instant Fiat Settlement) · On-Ramps and Off-Ramps Explained: How Crypto Becomes Usable Money