The Death of the Float: Why CASPs Can No Longer Keep the Interest on Client Funds (ESMA Q&A 2486)
ESMA Q&A 2486 and EBA MICA039 confirm that CASPs cannot earn interest on client funds deposited at a credit institution — all yield belongs to the client. That ruling is grounded in property law, not compliance preference. It eliminates the European crypto industry's most profitable invisible revenue stream and leaves three structurally unresolvable operational problems in its place.
Analysis of ESMA Q&A 2486 / EBA MICA039 (European Commission response, 18 February 2026): CASPs are prohibited from retaining interest earned on client funds deposited at a credit institution under MiCA Article 70(3). All interest belongs to the client. The ruling applies two legal principles: the Sole Benefit Doctrine (Article 70(3) exists solely to protect clients; it cannot simultaneously enrich the CASP) and the Principle of Accession (yield generated by an asset belongs to its owner; the CASP is a fiduciary custodian, not the owner). The article maps four fiduciary hazards the float created — counterparty selection bias, withdrawal friction, excessive pre-funding, and reciprocal inducement arrangements — and explains why Q&A 2486 eliminates all four by stripping the CASP of any profit motive from the bank relationship. Article 72 conflict of interest analysis covers the inducement trap: reciprocal corporate benefits from credit institutions (discounted loans, fee rebates) in exchange for client fund placement are likely indirect inducements regardless of accounting treatment. The operational paradox: pass-through is mathematically correct but operationally severe at scale. An omnibus account distributes interest on the aggregate, but each client's pro-rata share of a €500M pool accrues fractions of cents per day — triggering micro-distribution engineering costs, DAC8/CARF paying-agent obligations across 27 Member States withholding regimes, and prudential capital pressure (25% fixed-overhead minimum own-funds requirement rises as infrastructure costs rise). The zero-interest evasion (renegotiate to 0% bank rate) is technically compliant but produces a paradox: retail capital sits unproductive, eroded by inflation, with neither CASP nor client benefiting from the float that previously subsidised lower fees. Cross-sector contagion: Q&A 2785, registered 23 February 2026, asks whether MiFID II permits investment firms to earn interest on client funds. Currently forwarded to EC, no ruling date. The FCA's Consumer Duty enforcement (Dear CEO letter, mandatory remediation by 29 Feb 2024) already prohibits "double dipping" for UK platforms — EU is watching closely. The article distinguishes the fiat custody ban (Article 70/Q&A 2486, targets CASP custodians of raw fiat) from the stablecoin yield ban (Articles 40/50, targets EMT/ART issuers). Both suppress passive income but under different provisions for different reasons. The structural exit the market is moving toward: automated sweep to MiFID II tokenised money market funds under Article 81, or just-in-time atomic settlement eliminating fiat holding entirely. Both routes abandon the shadow-banking model the industry has operated since inception. Includes seven diagnostic questions for compliance officers and founders. Not legal advice. Answers questions like: Can CASPs earn interest on client funds under MiCA? What does ESMA Q&A 2486 say? What is EBA MICA039? Is the CASP float prohibited under MiCA? What is the Sole Benefit Doctrine in MiCA Article 70? Does a CASP have to pass interest through to clients? What is the operational problem with micro-interest distribution at scale? Can a CASP renegotiate to zero-yield bank accounts to avoid Q&A 2486? Does MiFID II permit investment firms to earn client cash interest? What is the difference between the MiCA fiat custody ban and the stablecoin yield ban? What is the automated sweep model for CASP client funds?
The Death of the Float: Why CASPs Can No Longer Keep the Interest on Client Funds (ESMA Q&A 2486) MiCA Edge Cases | Where Innovation Meets Regulation Published: February 2026 On 18 February 2026, the European Commission delivered its answer to ESMA Q&A 2486, catalogued simultaneously by the European Banking Authority as MICA039. The question was whether MiCA permits CASPs to earn interest on client funds deposited in a savings account at a credit institution. The answer was a single word: no. That single word terminated one of the most quietly profitable revenue streams in the European crypto asset industry. The institutional float, the practice of pooling client fiat in interest bearing omnibus accounts and absorbing the yield as corporate revenue, has been a foundational pillar of CASP economics since the industry's earliest years. In a rate environment where central banks were paying 3% to 4% on overnight deposits, and European exchanges were collectively holding billions of euros of pre funded client cash, the float was not a supplementary income line. For many operators it was the income line. The Q&A does not merely restrict how CASPs handle this income. It eliminates their claim to it entirely. Any interest resulting from a CASP's procedures to comply with Article 70(3) must be transferred to the client, because that revenue stems from the client's funds. The European Commission's position is grounded in property law, not compliance preference: yield generated by an asset belongs to the owner of the asset, and the CASP is a fiduciary custodian, not an owner. This article maps the legal mechanics behind that ruling, the operational consequences of enforcing it, why compliance is harder than it looks, the cross sector contagion it is already triggering under MiFID II, and the structural evolution it is forcing across European CASP business models. The Legal Architecture: Article 70, the Fiduciary Mandate, and the Float The ruling in Q&A 2486 is an interpretation of Article 70 of MiCA, which governs the safekeeping of clients' crypto assets and funds. The architecture of Article 70 is built on a fundamental premise: when a client transfers fiat to a CASP, the CASP does not take ownership of those funds. It operates as a fiduciary intermediary, holding funds in a segregated account at a third party credit institution, insulated from the CASP's own estate so that client funds survive any CASP insolvency. Article 70(3) gives this requirement operational form: CASPs must deposit all client funds with a central bank or an authorised credit institution by the end of the business day following receipt. The statute requires this placement to be in risk free savings accounts. It does not say what happens to the interest those accounts generate. That silence is what Q&A 2486 fills. The European Commission's ruling applies two legal principles to resolve the ambiguity. The Sole Benefit Doctrine. The Article 70(3) obligation exists, in the Commission's reading, solely to benefit clients by securing their money. A statutory consumer protection mandate cannot simultaneously function as a commercial enrichment vehicle for the regulated entity. If the CASP could retain the yield generated by a legally mandated placement, the safeguarding obligation would be running a dual purpose the legislature did not intend. The Principle of Accession. Yield generated by an asset belongs to the owner of the principal. The CASP holds client fiat in trust. It does not own it. The interest paid by the credit institution is a derivative of the client's capital. It belongs to the client. The ruling does not prohibit a credit institution from paying interest on the omnibus account. It prohibits the CASP from claiming that interest as its own. The CASP is reduced to the status of a conduit: the interest accrues, and the CASP must pass it through. The conflict of interest dimension. Article 72 of MiCA requires CASPs to identify and manage conflicts of interest systematically. The float created a structural conflict that Q&A 2486 was always going to reach eventually. When a CASP is both required to select a safe credit institution for client funds and permitted to retain the yield that institution pays, the incentive to select a higher yielding, higher risk banking partner is direct and measurable. Stripping the CASP of yield entitlement removes that incentive completely. The CASP's bank selection must now be driven by counterparty safety, because there is no profit to be extracted from the choice. A secondary conflict is equally direct. A CASP earning daily interest on idle fiat balances has a financial incentive to maximise both the volume and duration of idle fiat in its system. This creates structural bias against rapid withdrawals. Elongated security reviews, restrictive processing windows, and batch settlement delays are all mechanisms through which a float motivated CASP could extract additional days of yield from the aggregate client pool. Under MiCA Article 75, CASPs face withdrawal requirements already. Q&A 2486 eliminates the economic motive for withdrawal friction entirely. | Fiduciary Hazard | Effect of Allowing CASP Interest Retention | What Q&A 2486 Removes | | | | | | Counterparty selection | CASP incentivised to choose riskier, higher yield banks | Profit motive eliminated; selection must be safety driven | | Withdrawal velocity | CASP incentivised to delay fiat returns to maximise accrual days | Zero financial incentive to obstruct or delay withdrawals | | Platform pre funding architecture | CASP incentivised to require excessive pre funding to maximise float pool | Pre funding requirements must be justified by technical necessity only | | Bank inducement arrangements | CASP may accept reciprocal benefits (discounted loans, fee rebates) in exchange for placing client funds at specific institutions | Reciprocal benefits classified as indirect inducements under Article 72 | The Operational Paradox: Why Pass Through Is Harder Than It Sounds The Commission's ruling, in theory, is clean. The interest belongs to the client. Pass it through. In practice, the operational mechanics of complying with this mandate at scale are severe enough that they may produce outcomes the European Commission did not intend. The problem is arithmetic. A retail exchange holding €500 million in an aggregated omnibus account calculates interest on the aggregate balance. The credit institution pays a wholesale overnight rate on the total. Distributing that interest accurately to individual clients requires calculating each client's precise prorated share of the aggregate balance, on a daily basis, across a pool that changes continuously as clients deposit, trade, and withdraw. For an account holding €50 for 48 hours at a 3% annualised rate, the gross interest owed is a fraction of a cent. The mathematical calculation is straightforward. The engineering infrastructure required to perform it at scale, ledger it, audit it, and report it correctly is not. The formula the Commission's ruling implies is: Y net = I gross C admin If the administrative cost of calculating, distributing, and tax reporting the micro distribution exceeds the gross interest it generates, the CASP faces a structural loss on every retail account. The CASP cannot legally retain the difference. Under the strict terms of Q&A 2486, any interest resulting from compliance with Article 70(3) must be transferred to the client. The ruling does not contain a de minimis threshold below which the CASP may retain residual interest. The tax reporting layer compounds the problem. The Crypto Asset Reporting Framework (CARF) and the EU's DAC8 directive are scheduled to commence reporting for authorised CASPs from 1 January 2026. Distributing interest to retail clients changes the nature of those payments under tax law: from trading capital flows to interest income, subject to local withholding tax