In March 2025 we published an opinion piece arguing that the MiCA Regulation (Markets in Crypto-Assets), the EU’s first comprehensive rulebook for the sector, was suffocating the European crypto market: a dozen licences issued, thousands of national registrations about to expire, capital and talent drifting to the United States and Asia. Eighteen months later the MiCA transitional period has closed, the ESMA register of crypto-asset service providers has filled up, Binance has been pushed out of the EU and Tether never came in. The forecast can now be checked against the facts. This is the revised version of that article: what we said then, what actually happened, who won, who lost, and what still looks wrong with the European approach.
Editor’s note
The original text was written in March 2025 on the basis of Coincub’s Europe Crypto Report 2025. It was revised in September 2026, after the last MiCA grandfathering window expired on 1 July 2026. Figures from early 2025 are kept only where they mark the starting point; everything else reflects the current state of EU crypto regulation.
March 2025 against September 2026: the scorecard
Opinion pieces rarely get audited by their own authors. This one deserves it, because several of the numbers we relied on have moved a long way, and one or two were simply wrong. The table compares the European crypto market as we described it eighteen months ago with the position today.
| Indicator | March 2025 (as we wrote it) | September 2026 (as it stands) |
|---|---|---|
| CASP licences (authorised crypto-asset service providers) | 12 | 343 in the ESMA register as of 11 September 2026; roughly 244 before the 1 July deadline, then a rush of post-deadline approvals |
| E-money token issuers | 10 | 22, and still not a single asset-referenced token issuer anywhere in the EU |
| National VASP registrations | 3,167 across the EU, 75% expected to lapse | All lapsed. The final grandfathering window closed on 1 July 2026 and ESMA refused any extension |
| Largest exchange and largest stablecoin | Binance operating on national VASP registrations; USDT delistings only beginning | Binance withdrew its licence application and halted new EU services on 1 July; USDT delisted for EEA users because Tether never applied |
| Digital euro | Presented as imminent (our text even named December 2024) | Still not law: Council position December 2025, Parliament position July 2026, trilogue under way; pilot planned for late 2027, first issuance 2029 at the earliest |
| Venture capital for European crypto startups | $5.7 billion peak in 2022, down about 70% since | Roughly $1.5 billion of blockchain funding in 2025, almost half of it in Switzerland, outside the EU |
| United States | Strategic Bitcoin Reserve just announced | GENIUS Act on stablecoins signed in July 2025; CLARITY market-structure bill still stuck in the Senate |
Who won and who lost under MiCA
When we wrote that only 12 crypto-asset service providers held a MiCA licence, the number was accurate and alarming. Three months into the regime several member states had not even designated a competent authority, and we concluded that the CASP licensing machine was broken. In hindsight it was not broken, it was slow. The ESMA register held 199 firms in April 2026, about 244 by the end of June, and then absorbed 37 approvals in a single update on 3 July as regulators cleared applications that had been pending for a year. By mid-September the count stood at 343.
The winners: banks, payment institutions and the top tier of exchanges
Read the register and the names tell the story: BBVA, a BNY Mellon subsidiary in Belgium, Banking Circle, Bison Bank, Ripple Payments Europe, Bybit, Kraken, Coinbase, OKX, Bitpanda, Crypto.com. Germany alone accounts for 69 authorisations. The MiCA register reads like a banking directory with the largest crypto exchanges appended, which is exactly what our central thesis predicted: MiCA rewards balance sheets and compliance departments, not founders. What we got wrong was the scale. The market did not shrink to a dozen survivors; it consolidated into a few hundred well-capitalised crypto firms with EU-wide passporting rights.
The losers: Binance, Tether and the long tail
The most searched question of the summer was whether Binance had a MiCA licence to operate in the EU. It did not. After a Reuters report that the Greek regulator was preparing to reject its application, Binance withdrew the Greek bid on 24 June 2026, six days before the deadline, and halted sign-ups, deposits and new orders for EU users from 1 July while promising to apply in another member state. Tether never applied at all, so USDT was delisted for EEA retail clients on the major venues and the MiCA-compliant stablecoin market in Europe went to Circle. The lesson is not about size. The world’s largest exchange failed the same test as a two-person VASP: fit-and-proper ownership, governance and AML history. MiCA does not reward being big. It rewards being like a bank.
What happened on 1 July 2026
ESMA confirmed on 17 April 2026 that there would be no extension of the MiCA deadline, and on 23 June it told every unauthorised provider to wind down its EU activities. The Netherlands, Finland, Latvia, Hungary, Slovenia and Poland had already closed their national grandfathering windows after six months, Sweden after nine, Germany, Ireland, Lithuania, Austria and Slovakia after twelve. Poland became the cautionary tale: its crypto-assets act was vetoed by the President twice, in December 2025 and February 2026, and a third version was still working through parliament when the deadline hit, so Polish VASPs had no domestic route to CASP authorisation and could only survive by licensing elsewhere and passporting back in. In Estonia the FIU-era registrations are gone and authorisation by Finantsinspektsioon is the only lawful basis for serving EU clients; the post-authorisation obligations are now the daily reality for every provider that stayed.
MiCA compliance costs: the second filter comes in 2027
The July deadline tested one thing only: the ability to obtain a crypto licence in the EU. The first full financial year under supervision tests the ability to pay for one. A CASP authorisation is not a project with an end date but a permanent cost centre: a compliance officer and an MLRO, DORA-grade ICT risk management, annual audits, supervisory fees, own funds sitting idle, and a reporting calendar that runs whether or not there is revenue to report. These are fixed costs, and fixed costs punish small revenue. In July 2026 the head of Gate’s European entity said publicly that a good number of licensed firms would not be able to carry the ongoing compliance costs and resources that a MiCA licence demands. That is a licensed competitor describing the field, not a lobbyist arguing against regulation.
Three forces make 2027 the year the costs bite. EU passporting lets a thirty-country platform compete for local clients at zero marginal licensing cost, so the domestic moat that justified many small licences is gone. ESMA’s peer review of the Maltese authority in July 2025 found one authorisation granted with governance, ICT and AML issues unresolved and told every national supervisor to apply the lesson; the first reviews of the firms cleared in the pre-deadline rush fall in 2027. And the UK opens its own authorisation gateway on 30 September 2026, with applications accepted until 28 February 2027 ahead of a full regime from 25 October 2027, so a firm that has already built a MiCA-grade compliance function has somewhere to take it. Our forecast, to be checked in a year: the register will still grow in headcount, but the column of ended and withdrawn authorisations will grow faster; at least one exchange licensed in the first half of 2026 will have sold or surrendered its EU authorisation; and most new entrants in 2027 will be banks and payment institutions rather than crypto-native firms.
Crypto banking in the EU: still the hardest problem
In early 2025 Coincub reported that only 14% of crypto startups managed to open a bank account without a subsequent closure. Nothing in the regulation obliges a credit institution to serve a CASP, and debanking of crypto companies has not gone away. What did change is the mechanism. In spring 2026 banks began writing to clients that held only a national VASP registration, informing them that accounts would be closed on 1 July unless the client could show a pending or granted CASP application. The licence became the ticket to a bank account. In other words, the banking blockade turned into a licensing filter: a firm that could not afford the authorisation lost its account anyway. This is the one section of the original article we would not change a word of.
Where the venture capital and the jobs went
Europe’s share of global venture funding fell to 13% in 2025, from 16% a year earlier. For crypto and blockchain startups specifically, the CV VC Top 50 report puts Switzerland’s Crypto Valley at $728 million across 31 deals in 2025, which it describes as 47% of all European blockchain funding. That implies a European total of roughly $1.5 billion, still more than 70% below the 2022 peak. The detail that matters is geography: nearly half of the capital that stayed on the continent chose the one jurisdiction that sits outside the EU and outside MiCA, and Dubai licensing advisers reported a surge of enquiries from European founders in the weeks around the deadline.
Our original text quoted a collapse from over 100,000 blockchain vacancies in 2022 to about 10,000 in early 2025. The exact figure depends heavily on which job board is counted, so we no longer attach a number to it. What the H1 2026 hiring data shows instead is a change in composition. Among roughly 2,900 active crypto postings worldwide, engineering accounts for 34% and compliance and legal roles come second at 10%, while centralised exchanges and stablecoin or payment firms generate almost half of all demand. In Europe the new crypto rules have created a hiring wave for AML officers, regulatory counsel and reporting specialists. Whether that counts as a recovery depends on whether you are a lawyer or a developer.
USDT, euro stablecoins and the digital euro
Two facts about stablecoin regulation sum up what MiCA achieved and what it did not. First, removing Tether from regulated EU venues did not create euro liquidity; it handed the European market to a dollar token with a French licence. Second, MiCA-compliant euro stablecoins amounted to roughly $674 million in July 2026. That is growth of 128% in a year, and it is still about 0.2% of a $300 billion market dominated by dollar tokens. The EU regulated euro stablecoins into existence, but not into relevance.
On the digital euro we have to correct ourselves. The original article stated that the ECB would roll out a central bank digital currency by December 2024 and that it would cost €20 to 30 billion a year. Neither was right. The Council adopted its position on 19 December 2025, the European Parliament on 9 July 2026, and trilogue negotiations are running with adoption targeted for the end of 2026. The ECB plans a twelve-month pilot from the second half of 2027 and aims to be ready for a possible first issuance in 2029; its own estimate for building the system is about €1.3 billion. We stand by the substance of our criticism: a CBDC with holding limits and state-designed privacy tiers still does not answer a problem that European users actually have. But the numbers we quoted were wrong, and the ECB’s published timeline is the one to rely on.
MiCA against the GENIUS Act: two orders of operations
When we wrote in March 2025, the Strategic Bitcoin Reserve had just been announced by executive order. Since then US crypto regulation has done what Europe did in 2023, but in the opposite order. The GENIUS Act, signed on 18 July 2025, gave payment stablecoins a federal framework with full reserve backing, with implementing rules finalised through 2026 and enforcement from January 2027. The CLARITY Act on market structure passed the House in July 2025 and the Senate Banking Committee in May 2026, and was scheduled for a procedural Senate vote on 15 September 2026, with no guarantee of passing this Congress. The EU shipped one unified rulebook first and is still waiting for the market to arrive; the US regulated the product that already had a market and left the rest for later. One result is visible above: dollar stablecoins now carry a federal charter, while euro stablecoins carry a MiCA white paper and a rounding-error share.
What we got right, and what we got wrong
Right: consolidation, cost and banking access. The CASP register is dominated by banks, payment institutions and the top tier of exchanges; the firms that survived are the ones that could fund a two-year authorisation and a permanent compliance function; access to accounts remained the sharpest constraint, and the licence became the price of admission. Right, too, that the digital euro is a political project looking for a use case.
Wrong: the licensing process was slow, not broken, and the register filled up once national authorities started working. Wrong: several dates in the original text, including a transitional deadline placed in June 2024 that was in fact 1 July 2026, and a digital euro launch date that was never real. Wrong: the cost of the digital euro, overstated by more than an order of magnitude. Wrong, or at least premature: the claim that Europe would lose its status as a blockchain and crypto hub outright. Europe lost the startup layer and kept the institutional one, which is a different outcome and, for a regulator, probably the intended one.
What we would tell a founder today
A national VASP registration is worth nothing. The only lawful basis for serving EU clients is a MiCA CASP licence from a member state whose authority actually processes applications, combined with the EU passport. Estonia, Lithuania, Germany and the Netherlands have working regimes; Poland does not. The licence is no longer just a legal requirement: for most banks it is the precondition for keeping an account open, and from 2027 MiCA compliance is a cost line that has to be covered by revenue rather than by the funding round.
Conclusion: Europe has a rulebook, but does it have a market?
Eighteen months on, the MiCA framework has delivered what its authors promised on paper: a single crypto licence for the whole EU, a single passport, a supervised market and the removal of unregulated providers, including the largest one. It has not delivered what the industry needed: bank accounts, early-stage capital, euro liquidity on chain, or a reason for a founder to incorporate in the EU rather than in Zug, Dubai or Delaware. The startups we predicted would leave have left, or never started. The institutions we predicted would win have won. Whether that is a success depends entirely on whether the goal was a safer market or a larger one.
Our position has not changed, only the evidence for it has grown. Europe needs a lighter entry tier for small providers, a statutory right of access to basic banking for authorised CASPs, and a euro stablecoin policy that treats private issuers as partners rather than as competitors to a CBDC that will not exist before 2029. Until then, MiCA will remain a well-drafted crypto regulation for a market that is largely somewhere else.
