What happens if an investor loses access to their tokenized securities?
With ERC-3643, an authorised agent can recover the holding. The agent moves the full balance from the lost wallet to a new wallet that the investor has linked to the same verified identity. Any frozen units stay frozen, and the recovery is recorded on chain.
Key facts
- Recovery is required by the standard: it “MUST have a recovery system in case an investor loses access to his private key”.
- Only an agent of the token can carry it out, using
recoveryAddress(). - The new wallet must already be a management key on the investor’s on-chain identity.
- The whole balance moves, including frozen units and a frozen-wallet status.
- The token emits
RecoverySuccess.
Why can’t a normal token do this?
A plain ERC-20 token has no owner of last resort. If the key is lost, the units stay in that wallet forever. For a security, the legal owner still owns the asset whether or not they can sign.
How does it work, step by step?
The investor adds a new wallet as a management key on their on-chain identity. The agent calls recoveryAddress() with the lost wallet, the new wallet and the identity. In the reference implementation the token then:
- registers the new wallet with the same identity and country;
- moves the full balance;
- re-applies any freeze;
- removes the lost wallet from the registry;
- emits
RecoverySuccess.
If the new wallet isn’t a management key on the identity, the call fails with “Recovery not possible”.
Does the investor need to be verified again?
Their identity doesn’t change, so existing claims still apply. The agent’s own procedures decide how to confirm that the person asking controls that identity.