The Deposit Drain That Isn't: MiCA, Stablecoins, and the Lever the ECB Keeps for Itself

Market Note · The Deposit Drain That Isn't · 2026

The float the ECB warns about, and the lever it holds

EUR billions. Euro-area overnight deposits, the base both would drain, stand at 9,628bn, nine times past the chart edge. The lever runs roughly 1,500x the float.

Euro-area overnight deposits (context)

9,628 bn EUR

Euro-stablecoin float (the warning)Digital-euro drain at a 3,000 EUR cap (the lever)

Source: MiCA-compliant issuer data (Jul 2026), Bruegel estimate, and ECB (Jan 2026). Author’s calculations.nikkhosravipour.com

Speaking in Rome on 17 July, before the Federation of Italian Cooperative Credit Banks, ECB Executive Board member Piero Cipollone warned that "if the use of stablecoins increases in the future, banks will also lose retail deposits."¹ The float behind the warning is small. Euro-denominated stablecoins stand at about 674 million EUR, with Circle's EURC accounting for 430 million of it; the total is up 128% year to date. Euro-area overnight deposits stood at 9,628 billion EUR in January.² The entire euro-stablecoin market thus amounts to about 0.007% of the deposit base it supposedly threatens, and to roughly 0.3% of the 230 billion USD dollar-stablecoin float. A drain of this size is a rounding error, and for 2026 bank funding it is a non-event. The interesting question is why the warning is so loud when the float is so small.

MiCA's stablecoin regime, in full force since 1 July, makes the timing of that warning pointed. Issuers of e-money tokens, the regulation's term for the currency-pegged stablecoins that serve as the stable settlement leg on crypto exchanges and on the blockchain, now require authorization. Every token must be backed one-to-one with reserves held in bank deposits and high-quality liquid assets, and paying holders any yield is banned. The rules cover every stablecoin offered in the EU, dollar tokens included. Non-euro tokens face the stricter treatment: once daily use as a means of payment exceeds one million transactions and 200 million EUR in value, MiCA requires the issuer to stop issuing.³ The law already contains the dollar rail. Cipollone's warning arrived 16 days after the containment took effect.

Cipollone's audience explains the warning's real object. Cooperative credit banks live on retail deposits, and the deposits they fear losing would flow toward dollar tokens rather than euro ones. Dollar stablecoins make up about 99% of the global float and, since the GENIUS Act of 2025, carry a federal framework in the United States along with a captive bid for US debt. Wherever crypto trades, they are the default settlement asset. The fear is dollarization through the back door: a euro saved in a dollar token weakens the deposit base and, at scale, the ECB's grip on transmission. Precedent cuts the same way. Washington has given its stablecoins a legal home; Brussels worries about becoming a rule-taker on the digital form of its own currency.

For euro stablecoins, the near-term deposit drain rounds to zero. An e-money token cannot legally pay interest; a bank deposit can. Whatever reason a household might have to swap one for the other, rate competition is ruled out by statute. And a euro that does leave a household deposit rarely leaves the banking system. Under the reserve rule it lands in the issuer's reserve, of which at least 30% must sit in euro-area bank deposits and the remainder in short-dated paper, much of it bank and sovereign issuance. The aggregate deposit level holds roughly steady while the composition shifts from retail deposits toward wholesale issuer funding. That recomposition still costs banks something, since wholesale funding is costlier and treated worse in liquidity regulation; the cost arrives as margin pressure while the deposit base stays put. Dollar issuers supply the contrast: Tether and Circle park their reserves largely in US Treasury bills and repo, so dollar-stablecoin growth pulls balances toward the US Treasury market instead of back into the banks the money left. Net outflow from the euro area equals the float times the share of reserves parked outside the euro-area banking system, a share MiCA keeps small without pinning it at zero. At 674 million EUR of float, even total leakage would amount to 0.007% of the overnight base.

One instrument actually can drain euro-area deposits at scale, and it sits in the ECB's own drafting folder. The digital euro, as currently designed, is a non-interest-bearing liability of the central bank, held directly with it. A euro moved into a digital-euro wallet leaves the banking system entirely; no reserve recycles it back. The drafted holding cap of 3,000 EUR per person, chosen for sitting near the average euro-area net monthly income, bounds the exposure, with anything above the threshold sweeping automatically into a linked commercial-bank account, the so-called waterfall. Multiplied across euro-area adults, the cap implies potential migration of about 1,000 billion EUR, some 15% of aggregate retail deposits.⁴ That figure is a ceiling, resting on the assumption that every adult holds the full 3,000 EUR, an adoption level no deposit migration has approached. Even a modest fraction of the ceiling still dwarfs the float in question; the lever the ECB controls runs roughly 1,500 times the size of the thing it warns about.

That calibration is live politics. The Commission published the final digital-euro framework on 11 July; the first Council-Parliament trilogue followed on 13 July, with the holding limit an open gap; the pilot is set for mid-2027 and first issuance targeted for 2029. The deposit event, if it comes, is years out and hinges on one political number. A higher cap widens the drain. A lower cap leaves the disintermediation trade dead on arrival. The ECB frames the product as protection for 300 million bank accounts, and that a central bank must cap its own instrument to deliver that protection is the clearest available statement of where the risk actually lives.

All of this is testable, and the tests worth running are dated. The first is the holding-limit calibration as trilogues run toward year-end: a limit set at the high end of the debated range opens the drain this note locates in the digital euro. The second is any movement on the e-money-token yield ban, whose repeal would restore the rate incentive and flip the stablecoin call outright. A third sits further out: a scaling of EURC by two orders of magnitude, which would still leave the euro float under 1% of the deposit base. The correlation test runs in the background; should euro-area household deposits ever fall in step with euro-stablecoin issuance, controlling for rates, the thesis here fails, though at the current float the effect would be too small to read either way. For positioning, the implication runs against the mood: a de-rating of European bank equity priced on stablecoin deposit flight, whether through deposit betas or net interest income, is mispriced and years early. Fading it beats chasing it.

The gap between the fear and the facts is the finding. The ECB warns about a 674 million EUR float that its own regulation has rendered incapable of draining deposits, while it designs a trillion-capable instrument and sets the dial itself. What the warning defends is the bank-deposit and credit-creation channel, and behind it the monopoly on the unit of account; the immediate triggers are the dollar tokens and the precedent they set. Hayek argued in 1976 that no monetary authority would allow currency competition to run far enough to threaten its position. The ECB now proposes to meet private money with a product of its own, capped so that the competition cannot go too far. The design concedes the argument.

Footnotes

Get new essays and research by email. Subscribe on Substack ↗ (opens in new tab)