France Is Not Taxing Stablecoins. It Is Reclassifying Them.

On October 16, the French National Assembly advanced a provision that would make stablecoin swaps a taxable event. The reporting contained six information points. It contained zero primary sources. No tax rate. No tax base. No amendment number. No vote tally. The claim that stablecoin swaps "take the hit" is the entire evidentiary record, and it is a direction, not a datum.

My first reflex, trained across three weeks of reconciling a leaked FTX ledger against public on-chain deposits, is to locate the missing variable. In accounting, the absent number is not an inconvenience. It is the finding. A tax provision that cannot be quantified cannot be evaluated; it can only be believed. So the honest opening is not "France taxes stablecoins." It is: France has moved to reclassify what counts as a taxable disposal, and has declined, so far, to say at what price.

France's current treatment is legible. Disposal of crypto for fiat is taxed under a single flat levy — the prélèvement forfaitaire unique, roughly 30 percent including social contributions — applied to realized gains. Crypto-to-crypto swaps have historically been treated as deferred: exchanging one asset for another does not, by itself, crystallize a taxable event. This deferral is the seam the proposal targets. If a euro stablecoin is legally an electronic money token under MiCA, then swapping into it is not a neutral rotation between digital assets; it is, functionally, a conversion close to fiat. The provision does not appear to introduce a new tax. It redraws the boundary of the old one.

MiCA distinguishes electronic money tokens, asset-referenced tokens, and other crypto-assets; only the first is named here, and that naming is the whole ballgame. This matters because the instrument is a budget bill. The provision is almost certainly an amendment to the projet de loi de finances — the annual finance act — and its fate is bound to the entire budget, not to an independent vote on crypto. France's recent budget politics are unstable; whole finance acts have been forced through under Article 49.3 or truncated by censure. A crypto clause riding inside that vehicle inherits all of its fragility. The October 16 first reading is the opening of a procedure, not its conclusion. The Sénat reviews next; a commission mixte paritaire reconciles the chambers; the Conseil constitutionnel audits the result; only then does promulgation set a January 2027 effective date.

Here is the operative insight, and it is not about rates: this is a taxonomy operation disguised as a tax increase. The legislator is not inventing a levy. It is selecting one legal category — the MiCA electronic money token — and reclassifying its acquisition-by-swap as a disposal. That distinction determines everything downstream, because it means the policy's success depends on classification clarity, not on enforcement aggression.

The choice of the EMT category is not accidental. Under MiCA, an EMT has a licensed issuer, a reserve-disclosure obligation, and an identifiable contract whitelist. A tax authority facing "any ERC-20" has no tractable perimeter; a tax authority facing a regulated token class has a registry. Executability has been designed into the statute. The French treasury is not chasing the entire on-chain universe; it is drawing a fence around the one segment that already reports to a regulator. Compliance, in this architecture, is the precondition for visibility — and visibility is the precondition for taxation. The MiCA seal that issuers marketed as legitimacy becomes the bullseye.

The effective date confirms the logic. January 2027 is not a round number chosen for ceremony. DAC8 — the EU's eighth directive on administrative cooperation — required member-state transposition by the end of 2025 and applies from January 2026, with its first reporting year covering 2026 data and delivery in 2027. The OECD's CARF framework schedules its first exchanges for 2027 as well. The proposal's activation aligns, to the month, with the arrival of exchange-reported data. The inference is unavoidable: first data, then enforcement. The algorithm remembers what the witness forgets. Legislators are not waiting for political will; they are waiting for the reporting rails.

That coupling also exposes the structural blind spot. If the taxable event is defined as an EMT swap, then the capture mechanism is centralized intermediary reporting. A compliant exchange in Paris will file. A decentralized exchange, a cross-chain bridge, or an offshore venue will not — because no intermediary exists to file. This produces a predictable asymmetry: centralized venues become tax collectors, self-custody becomes an escape hatch. The provision taxes the visible and misses the invisible, which is the inverse of good tax design and the precise incentive gradient that pushes activity toward the unmonitored edge. In my audit of three Optimistic Rollup bridges in 2024, I watched a single misclassified state transition permit infinite minting under a race condition; the lesson generalizes. A system's behavior is governed by its boundary conditions, not its stated intentions. Draw the boundary at "EMT swap," and rational actors will walk around it — into non-EMT stablecoins, into bridged wrappers, into jurisdictions that do not file.

Which raises the second technical fault line: cost basis. If each EMT swap is a discrete taxable event, then repeated rotation — routine in DeFi lending, liquidity provision, and bridge transit — triggers repeated taxation unless the statute carries a continuous cost-basis tracking mechanism and explicit exemptions for same-asset transit. Is USDC bridged from Ethereum to Arbitrum a disposal? Is wrapped EMT the same asset as native EMT? The reporting discloses no exemption clause. Without one, tax liability can exceed economic substance; the taxpayer can be taxed on a gain that never economically occurred, purely as an artifact of an unmaintained ledger. Ledgers balance, but ethics remain uncalculated. This is the single most likely point of amendment in the second reading, and the single most likely source of litigation if it survives.

There is a second-order commercial signal buried in the timing. A fifteen-month runway between passage and activation creates certain demand for on-chain accounting and tax-aggregation tooling; France already hosts crypto tax software firms, and the compliance layer will grow. But this is a derivative business, not a technical breakthrough, and it does not touch the underlying classification problem. It merely monetizes the friction the statute creates.

Two further risks follow from the missing parameters. First, horizontal equity: if the provision taxes EMT swaps while leaving ETH/BTC interchange deferred, two economically identical rotations carry different burdens. France's Conseil constitutionnel reviews tax equality with some rigor; the clause invites a constitutional challenge on precisely that ground. Second, and larger, the Laffer arithmetic: a sufficiently high rate on swaps does not raise revenue — it relocates the swap. France already competes against Germany's one-year holding exemption, Portugal's historically favorable treatment, and the near-zero capital-gains regimes of the UAE and Switzerland. A tax on the medium-of-exchange function of stablecoins does not touch the store-of-value function; long-term holders are untouched, while market makers, arbitrageurs, and DeFi users — the participants who actually keep on-chain liquidity alive — absorb the friction. Proof exists; it is merely waiting to be verified in the Sénat's redline, but the direction is clear.

The consensus dismissal is that this is not tradeable, and on that narrow claim the dismissal is correct. First reading plus a 2027 activation plus an EMT-only perimeter is a triple attenuation; the discount rate applied to a two-year-out conditional clause on a single token class leaves price impact in the low single digits. BTC and ETH do not reprice on this. But the crowd that stops there commits a reasoning bug: it treats a missing parameter as a zero parameter. The rate is unknown; the unknown is not the same as the benign. And the optimists who note the fifteen-month implementation window are reading it correctly — the runway is deliberate, a signal that legislators intend an industry adaptation period rather than a punitive shock. Where both camps miss is the signal layer. The durable information is not the clause; it is that a major EU treasury has begun systematically treating stablecoin swaps as a deferral channel to be closed. That intent will resurface in another form even if this text dies in committee.

The rate will arrive; the classification has already arrived. Track three nodes: whether the Sénat deletes the clause, whether the CMP revives it, and — the real fork — whether scope expands from EMT-only to all crypto-to-crypto swaps, which would convert a local friction into a French market-competitiveness event. Everything else is a rounding error until the amendment text is on the record. Read the primary source. The headline is a direction; the ledger is the truth.