Amundi Puts a 600-Stock European Index on Ethereum, and the Custody Question Gets Louder

The Numbers First

Europe's largest asset manager has tokenized a share class of an equity index fund on Ethereum. The fund tracks 600 companies across European markets. Amundi runs something close to €2 trillion, and that scale is the whole story here. A boutique doing this is a pilot. A €2 trillion manager doing this is a signal.

Now the part the headline skips. This isn't a tokenized stock. It isn't synthetic exposure. It's a fund share class whose ownership record is maintained, at least in part, on a distributed ledger. The legal claim is the same one a pension consultant has approved a thousand times over. Only the plumbing is new.

That matters more than the chain does.

No asset target for the share class has been published, and the silence is a tell. Tokenized fund classes launch small, prove the operational rails, then scale when a distribution partner signs on. The first €10 million is a technology test. The first €500 million is a business.

Why the Wrapper Beats the Token

Tokenization started in cash, and it started that way for a reason. Franklin's on-chain government money fund launched back in 2021 as the first US-registered mutual fund to use blockchain for recordkeeping. BlackRock's BUIDL arrived in March 2024 and cleared $500 million in roughly six weeks. Both were Treasury products. Short duration, clean corporate actions, one issuer behind every line item.

Equities are harder, and European equities are harder still. Six hundred companies means six hundred dividend schedules. It means six hundred sets of proxy voting instructions. It means a withholding tax matrix spanning every bilateral treaty on the continent, each with its own documentation requirements and its own reclaim process.

That's the actual work. It's an operations problem, not a cryptography problem, and it's the reason most tokenization projects die in the middle office rather than on the chain.

Amundi picked the difficult version on purpose. If the rails hold, expect copycats within four quarters. Asset managers watch each other's operational spend the way traders watch each other's fills.

What Allocators Ask Before Anything Else

The custody question remains the gating factor for most allocators. Who holds the underlying shares? Which entity acts as transfer agent, and does that entity have a supervisory blessing for ledger-based records? What happens if the chain halts, or forks, or the validator set changes hands?

And then the practical one. Can the token move to a wallet the manager doesn't control? On most institutional designs the answer is no, and that answer collapses a great deal of the marketing.

A permissioned share class on a private-adjacent ledger is, for most allocators, a database with extra steps. It becomes genuinely interesting at the moment it enables something the old rails can't. Same-day collateral mobility. T+0 settlement against tokenized cash. Use as collateral in a lending facility without a custodian in the loop for three days.

None of that's proven yet. Fiduciary obligations demand more than conviction. They demand process, and process means an operational due diligence questionnaire that most allocators haven't written yet.

Ethereum won this round by default, by the way. It's where the tooling sits, where the auditors have done the work, and where the liquidity pools. That's not ideology. Any allocator treating chain selection as a values statement is mispricing the risk.

The Tests That Actually Matter

Watch three things, and watch them on a calendar rather than a vibe.

The EU's DLT Pilot Regime expires in March 2026. That deadline forces a decision in Brussels about whether ledger-based settlement becomes permanent infrastructure or reverts to a sandbox that closes. MiCA has applied in full since December 30, 2024, so the token framework exists. The securities framework is the piece still in motion.

Second, watch for a second manager. One European equity index on Ethereum is an experiment. Two within six months is a category.

Third, watch the asset figure in the fund's semi-annual report. That's the honest scoreboard. If this share class clears €100 million inside four quarters, the model works and the queue forms behind it. If it's still a rounding error in the annual disclosure, the industry will quietly reclassify tokenized equities as a 2030 conversation and go back to Treasuries.

So what should an allocator actually do with this? Nothing, strategic allocation. Your policy portfolio doesn't change because a wrapper got cheaper to administer. But your operational diligence does, and it changes within two years rather than ten.

The risk-adjusted case for the underlying equity exposure is unchanged. The risk-adjusted case for holding it through a twenty-year-old settlement stack is starting to look like the weaker bet. The custody question remains the gating factor, and for the first time a €2 trillion manager has given its counterparties a reason to answer it.