July 1, 2026 was the hard close. The Markets in Crypto-Assets transitional period ended with no extension: ESMA confirmed in April there was no mechanism inside the regulation to grant one, and told unauthorized providers to stop onboarding, stop marketing, and wind down to orderly exits. A crypto-asset service provider without a MiCA licence from at least one member state on July 1 lost the right to serve any of the 27.
The headline number looked like a purge. Of more than 1,200 entities registered under the old national regimes, roughly 210 secured full MiCA authorization by the deadline. More than 80 percent of registered firms did not clear the bar. On paper, four out of five operators were shown the door.
That framing is wrong, or at least it measures the wrong thing. Counting firms treats a venue that clears a few million euros a year the same as one that clears a few billion. The question that matters for anyone actually trading is not how many licences were issued. It is where the volume sat before enforcement, and whether it moved after.
The concentration was already there
On the volume that counts, the cleanse changed very little, because the market was already consolidated before MiCA touched it. By Kaiko’s June 2026 read, exchanges holding a MiCA licence already accounted for about 83 percent of European trading volume. The licensed venues did not win market share on July 1. They walked in already holding it.
Zoom into euro-denominated spot and the picture sharpens further. Kaiko’s data across 2025 and 2026 puts Bitvavo at 44 percent of global EUR spot volume, Kraken at 20 percent, Coinbase at 13 percent, and Binance at 8 percent. Three of those four are licensed. Kaiko’s longer-run market snapshot has four platforms accounting for over 85 percent of euro-denominated volume well before the deadline arrived. MiCA did not build this structure. It ratified it, and handed the incumbents a passport that makes the moat legal as well as commercial.
Binance’s exit was thin, and orderly
The one large operator that did not clear the bar was the largest globally. Binance withdrew its MiCA application in Greece on June 24, six days before the close, after regulators reportedly focused on fit-and-proper questions and its anti-money-laundering history. It emailed users in France, Italy, Poland and Spain in the days before the June 30 cutoff. From July 1, EU clients lost new spot orders, deposits, sign-ups, and Earn and staking products.
Two details keep this from being a crisis. First, withdrawals stayed open, so there was no trapped-funds event, only reports of congestion and slow processing as French users pulled balances at once. Second, and more telling for market structure, euro trading was about 1 percent of Binance’s global spot volume. The exchange that dominates the global order book was a marginal player in euros. Its EU departure reshuffles the roughly 17 percent of European volume that sat outside the licensed cohort. It does not dent the 83 percent that already had the paperwork.
So the enforcement did not deliver a shock. It delivered a sorting. The licensed incumbents that hold the single-market passport, Coinbase, Kraken, Bitvavo and their peers, keep the deep euro books they already had. The unlicensed long tail is gone, and the one large exile was thin in the currency that MiCA actually governs.
The open question is retail, not volume
The volume story is settled. The migration story is not. Binance’s EU retail book had to go somewhere, and the destination decides whether MiCA achieved its stated aim or merely relocated the risk.
There are three paths, and they are not equal. Retail can move to licensed venues inside the perimeter, which is the outcome the regulation was written to produce and the one the concentration data quietly favors, since Bitvavo and Kraken already run the cheapest, deepest euro books. It can move offshore, to non-EU platforms that solicit European users through channels MiCA cannot police, which keeps the trading alive but strips the consumer protections the whole exercise was meant to buy. Or it can move into self-custody and decentralized venues, which is outside the licensing regime entirely.
None of the current data resolves this. The Kaiko concentration figures measure venue volume, not user counts, and a 1-percent-of-Binance euro book is small enough that even a full migration to licensed venues would barely move the 83 percent share. That is precisely why the number to watch over the coming quarter is not exchange market share, which is now close to locked, but euro on-ramp flow and licensed-venue account growth. If licensed venues show a step-up in new European accounts through the third quarter, MiCA captured the retail it displaced. If they do not, the retail left the perimeter, and the regulation cleaned up the registry without cleaning up the risk.
Thirty days in, the verdict on market structure is clear and slightly anticlimactic. Enforcement did not break the European market or trigger a liquidity event. It certified an oligopoly that already existed and made its boundaries legal. The interesting part was never the 80 percent of firms that failed. It was the 1 percent of Binance’s book that had to find a new home, and whether it stayed inside the fence MiCA just finished building.
Discussion
Sign in to join the discussion.
No comments yet. Be the first to share your thoughts.