Rehypothecation, explained — and how to prove it never happens

Rehypothecation is when a custodian, broker, or lending platform re-uses an asset a client has pledged to it. The client still owns the asset on paper. The custodian, meanwhile, posts it as collateral for its own borrowing, uses it to fund another position, or lends it out to earn yield.

In traditional finance this is normal and, within limits, legal. Prime brokers rehypothecate client collateral every day, under written agreements and, in jurisdictions like the US, regulatory caps (the SEC’s rules limit it to 140% of a client’s debit balance) that disclose how much can be re-used and on what terms. The client knows it is happening and is compensated for it.

In crypto, the same practice has often happened without disclosure, without limits, and without reserves set aside to make clients whole. That is the version that hurts people, because the asset can be gone long before anyone is told.

The problem with rehypothecation is rarely the practice itself. It is that you usually cannot see it on a statement. Your balance reads the same whether the asset is sitting untouched or has been lent out three times.

A few common hiding places:

• Omnibus accounts. When client assets are pooled into one commingled wallet, individual ownership is tracked on an internal ledger, not on-chain. There is no external way to confirm your specific coins are still there.
• Lending desks. A custodian with an in-house lending arm can move pooled assets into loans without changing what any single client sees.
• “Yield” products. If a custody or exchange product pays a return, that return has to come from somewhere. Often it comes from putting customer assets to work, which is rehypothecation by another name.

The statement looks unchanged because the statement is an internal record, not a proof. The asset and the accounting have been separated.

Rehypothecation is hard to detect from the outside precisely because nothing visible changes. The shortfall only becomes real when too many clients ask for their assets back at once and the custodian cannot return what it does not have.

This is the pattern behind several large crypto failures. Customer assets were treated as a funding source, the re-use was not disclosed or reserved against, and the gap stayed hidden until a redemption wave forced it into the open. By then the assets had already been spent or lost.

The fix is structural, not contractual. A clause promising not to rehypothecate is only as good as the audit that checks it, and that audit happens after the fact. The stronger approach is an architecture where re-use is not possible and where you can verify the state of your assets yourself, at any time.

Four properties make rehypothecation provably absent rather than merely promised:

• Segregated per-client accounts.Assets held in their own on-chain UTXOs, not pooled into an omnibus wallet, so each client’s holdings are distinct and individually verifiable.
• On-chain proof of reserves. The ability to confirm directly on the ledger that the assets backing your position exist and are where they should be.
• Keys under your control. Multisig and hardware or HSM signing arrangements (for example 2-of-2 or 2-of-3) where the custodian cannot move assets unilaterally. No single party can pledge what it cannot move alone.
• No rehypothecation by design. Re-use ruled out by how the system is built, not by a paragraph in a contract.

On the Liquid Network, Blockstream’s custody approach combines these. Each client’s assets sit in segregated UTXOs rather than a commingled pool, so there is no shared balance to quietly draw down. Reserves can be confirmed on-chain, and Confidential Transactions let a holder prove balances to an auditor or regulator through selective disclosure without exposing them publicly. Signing requires keys the client controls. Put together, there is no point at which a custodian can re-use an asset without the client’s own key, which means rehypothecation is not a risk to be policed. It is a state the system cannot reach.

If you hold assets with a custodian, broker, or exchange, these questions surface the difference between a promise and a proof:

• Are my assets held in a segregated account, or pooled in an omnibus wallet?
• Can I verify my reserves directly on-chain, on demand, rather than relying on a periodic report?
• Can you move my assets without a key that I hold?
• Does any yield or lending program touch my assets, and if so, is it disclosed and reserved against?
• Is no-rehypothecation a contract term, or a property of how custody is built?

If the answers depend on trust and after-the-fact audits, rehypothecation is possible. If they depend on segregation, on-chain verification, and keys you control, it is designed out.