One Year Ago, Germany Gave EU Holders of USDe 42 Days to Claim, Then Moved Their Claim Offshore. Here Is What the Exit Looked Like, and How to Check Any Synthetic Dollar's Exit Path Before You Need It.

A year ago today, on August 6, 2025, a clock ran out. It was the last day German and EU holders of Ethena's USDe could assert a redemption claim against Ethena GmbH, the German entity that BaFin had ordered to wind down. From the next morning, August 7, 2025, those claims could be asserted only against Ethena's British Virgin Islands entity, Ethena (BVI) Limited, with USDe redeemable there for USDC. It was the first redemption BaFin had ever forced under the EU's crypto rules, and it went about as smoothly as a forced exit can: the peg held, a BaFin-appointed representative oversaw the process, and holders who filed got USDC.

We wrote about the regulatory side of this at the time, the way BaFin pushing USDe out of Germany and BlackRock pulling it into its institutional platform could both be right, in BaFin Barred USDe. BlackRock Integrated It.. This is not that piece. A year on, the more useful thing to study is not who was right about USDe. It is the exit itself, because what happened to EU holders is the thing most holders of any synthetic dollar have never once checked for their own: what the exit path actually is, who controls it, and whether it is a right you hold or a permission you are granted.

What a forced redemption actually looked like

Start with what did not happen, because it is the part people expect. USDe did not collapse. It did not depeg. There was no bank run and no haircut. USDe's supply is public and its price is readable, and through the window it held its dollar peg; holders who redeemed received USDC roughly one for one. By the standard of the collapses in our graveyard of dead synthetic dollars, this was the good outcome. The instrument worked.

Now the part that is easy to miss precisely because nothing broke. The exit was not on the holder's terms. It ran on a 42-day clock the holder did not set. It required filing a claim with a specific legal entity, and on a fixed date that entity changed: after August 6, 2025, the counterparty a holder had a claim against was no longer the German company they had dealt with but a BVI company most of them had never heard of. And the asset you redeemed into was chosen for you: USDC, not necessarily the thing you would have picked. None of that is a scandal. It is simply what it means for an exit path to be controlled by the issuer and the regulator rather than by you. The value survived. The control was never yours to begin with, and the wind-down is just when that became visible.

The exit path is the thing you check last and need most

Here is the pattern worth internalizing from the anniversary. Holders spend their diligence on the yield and, if they are careful, on the backing. Almost no one checks the exit path, the actual mechanics of how you get your money back out, until the day they are made to use it. That is exactly backwards, because the exit path is the part that decides what happens to you in the one scenario diligence is for: the bad day.

An exit path has three properties, and each one is checkable in advance. Ask them of any dollar, synthetic or otherwise, before you hold it, not after.

1. Can you redeem directly, or only sell into a pool? A dollar you can redeem with the issuer for the underlying has a floor. A dollar you can only exit by selling into secondary-market liquidity has whatever the pool has, which on a bad day is nothing. Magic Internet Money is the cautionary case: with redemption effectively unavailable, holders were left selling into a one-sided pool, and MIM traded near fourteen cents while its collateral sat readable on-chain, as we walked in the MIM depeg teardown. Direct redeemability is the difference between an exit and a fire sale.
2. Who holds the redemption switch, and can they pause it? If redemptions can be halted by a single party, an off-chain operator, a multisig, a regulator, then your exit is a permission that party grants, not a right you hold. That is not always wrong (a pause can protect holders in an incident), but you should know it before you need the exit, because a permission can be withdrawn at exactly the moment you want to use it.
3. Is it one to one for a hard asset, or a floating redemption value? Some dollars redeem one to one for USDC or the underlying. Others redeem at a floating figure the issuer computes, which is explicitly not a fixed dollar. Apyx's apxUSD, which we covered in the apxUSD read, redeems at a stated Redemption Value that tracks a basket rather than a peg, and characterized trading below a dollar as expected behavior of the model. Neither design is wrong, but they are different promises, and the exit is where the difference lands on you.

The Ethena wind-down scored well on the first of these (holders did get redeemed, one for one, into USDC) and it is the second one, who controlled the switch and the clock, that the anniversary illustrates. The holders were fine. They were also, for 42 days, entirely dependent on a process someone else ran.

Where Kerne stands on its own exit path

kUSD owes an honest answer to the same three questions, including where the answer is only partial. On the first, kUSD is directly redeemable: the Peg Stability Module holds USDC one to one against the kUSD it mints, and the redeem path returns USDC from those on-chain reserves, so a holder is not forced to sell into a secondary pool to exit. The caveat belongs in the same breath: the reserve sits across three PSM contracts, a live mint PSM and two retired ones kept as redeem-only reserves, and each pays only from its own balance, so a redemption larger than the deepest leg has to be split across sequential transactions rather than filled in one. On the second, the redeem path is an on-chain contract rather than an off-chain desk, so it is not a permission a single operator grants day to day; the honest caveat is that the multisig can flip a kill switch in an incident, and we would rather you know that from this sentence than discover it in one. On the third, the PSM redeem is one to one for USDC, not a floating redemption value.

And the boundary, stated in the same breath as everywhere else on this site: kUSD is one external audit old and at Genesis scale, and the deployed vault runs earlier bytecode than the commit Hexens reviewed, so those findings are live on chain; the delta-neutral hedge sits on Hyperliquid as a single venue, that leg is self-reported rather than on-chain-recomputable, and it is not running against the collateral that backs kUSD today, which is why our own signed attestation publishes delta_neutral false and a WARNING_DELTA status; and the full set of conditions under which the protocol would wind down or pause is written out, in advance and in public, in the exit triggers and emergency runbook. A wind-down should be a document you can read on a calm day, not a surprise you meet on a bad one. That is the difference the anniversary is really about.

Check the exit path on anything you hold

The skill this anniversary is a reason to practice applies to every dollar in your wallet, USDe included. Before the next forced exit, run the three questions, and use the on-chain reads to answer them rather than the marketing page.

• Read the redeemable reserve, not the claim. /verify-anything reads the live on-chain supply of any stablecoin and its reserve where that is on-chain, so you can see whether the backing you would redeem against is actually there, and readable, before you need it.
• Find the switch. For any dollar, ask who can pause redemptions and where that authority lives. For kUSD the authority surface is public at /api/risk-status and rendered at /risk; for others, it is the diligence a counterparty verification is for.
• Read the wind-down terms in advance. If a dollar publishes what happens to holders in an exit, read it now. If it does not, that absence is itself the answer to the question.

One year on

The Ethena EU wind-down is not remembered as a disaster, and it should not be, because holders came out whole. The reason to mark the anniversary is quieter than a disaster and more useful: it is the clearest recent proof that a synthetic dollar's exit path is a real, load-bearing thing, controlled by someone, on terms set before you ever go to use it. Check yours while nothing is wrong. The exit is the one part of diligence you do not get to do over.

Figures and status are as of the publication date, and nothing here is investment or legal advice, nor a claim that USDe is under-backed. BaFin's wind-down of Ethena GmbH, the June 25 to August 6, 2025 redemption window (42 calendar days), and the transfer of claims to Ethena (BVI) Limited with redemption for USDC from August 7, 2025, are per BaFin's own published June 25, 2025 notice; the BVI redemption is the ongoing fallback and there was no separate 2026 redemption deadline. USDe supply and price through the window are readable at DefiLlama. The MIM and apxUSD exit-mechanics references are drawn from our own dated teardowns linked above. Kerne's claims resolve to live endpoints: the redeem path and reserve at /api/por, the authority surface at /api/risk-status, and the wind-down conditions in the exit runbook. A /verify pass proves an attestation is authentic and fresh; it is not an audit and not a solvency opinion.