Why the EU Just Told Polymarket and Kalshi to Get Out — and Why It's Not Really a MiCA Story

The headline billed it as a MiCA story, but the more useful reading of why Europe just told Polymarket and Kalshi to get out is a fight over a category — and the category decides everything.

On September 10, 2026, the European Securities and Markets Authority warned that the two biggest prediction-market platforms do not hold the authorization required to sell event contracts to EU users. That ran alongside an ESMA risk report that called the sector rife with insider trading. The two companies were the ones caught in the crosshairs, but the mechanism applies to the whole industry, and it is worth unpacking because the press shorthand is genuinely misleading.

The category floor is older than crypto

Nothing about this starts with MiCA. In 2018, ESMA used emergency powers to ban firms from marketing binary options to retail investors across the EU — a product that pays a fixed amount if an event happens and nothing if it does not, which is exactly what a prediction-market contract is. Last July ESMA issued a public statement making the bridge explicit: event contracts with binary payouts tied to financial instruments are derivatives under the trading rules, so the retail binary-options ban covers them. That single step is the load-bearing fact. It means the EU has no accommodation for the product to be sold to ordinary retail consumers, license or not.

That is why the count of licensed prediction-market operators in the EU is zero, and why there is no obvious path to a license. Building one means satisfying a web of rulebooks at once — the securities framework for contracts tied to financial instruments, national gambling law for contracts that are not, and MiCA only for tokenized event contracts that dodge the derivative label. Polymarket, which settles bets in USDC on Polygon, sits inside that seam: a product that looks like a derivative in some cases, gambling in others, and a crypto asset in still others, with no single designation that gets it through the door.

The walls were already going up before the official warning

The September statement formalizes a squeeze that had already been building at the member-state level through 2026. Spain's gambling regulator issued a precautionary block in late May. Nine European gambling regulators released a joint warning in June. France blocked Polymarket at the ISP level in July, and restrictions piled on in Belgium, the Netherlands, the Czech Republic, Romania, and Portugal. The pattern is a member-state groundswell meeting an EU-level confirmation — not a rumor.

What this does to the investment thesis

Neither Polymarket nor Kalshi is publicly traded, so a US retail investor cannot buy the stock. The relevance is to the sector's valuation story and to anyone using the platforms. Through 2026 the private markets went from treating prediction markets as a curiosity to pricing them like a breakout asset class: Kalshi confirmed a $1 billion round at a $22 billion valuation in May after climbing from roughly $5 billion in early 2025, and Polymarket — the world's largest prediction market, the platform that carried the 2024 US election boom — was raising at around $15 billion and now runs its US business as a CFTC-regulated exchange.

Those numbers assume a global, mass audience. The EU removes that assumption for roughly 450 million people, and it is explicit that no licensed operator exists and that authorization would take well over a year with no precedent to lean on. That is not a temporary compliance lag; it is a structural ceiling on the total addressable market, set by a 2018 product ban the sector never had a lobbyist seat to repeal.

There is a second-order point worth keeping. The EU's insider-trading worry is not a regulator being squeamish; it is the economics of prediction markets biting themselves. A prediction market's whole value proposition is that it rewards people who know something to bet on it — but that incentive is the same reason regulators classify the product as something close to a banned security. The feature the bull case sells is the feature the law treats as the defect.

Verdict: the MiCA framing dressed up a category problem as a licensing problem. The category problem is that a product the EU banned for retail back in 2018 has since grown into a $20-billion valuation without any EU home — and the category, not the license, is the hard wall. For a US retail reader, the practical lesson is symmetrical with the US side: the moment a jurisdiction decides what category these contracts belong to, it decides how fast they can grow. Europe just answered, and the answer caps the audience. Whoever clears a real EU path first — or gets the category reclassified — is doing more for the sector than any license filing.