Proof of Reserves

Proof of reserves is a method by which an exchange or custodian demonstrates that it actually holds the assets it owes its customers. The custodian cryptographically proves control of on-chain wallets holding its reserves and publishes a commitment to total customer balances, letting outsiders check that assets cover liabilities rather than taking the company's word for it.

A typical implementation has two halves. On the asset side, the exchange signs messages from its wallet addresses, proving it controls them. On the liability side, it builds a Merkle tree of all customer balances and publishes the root; each customer can verify that their own balance is included in the total. If assets equal or exceed the proven liabilities, customer funds are at least nominally backed one-to-one. For example, after the FTX collapse in 2022 revealed that customer deposits had been quietly lent out and lost, many major exchanges began publishing regular Merkle-based proof-of-reserves reports so users could verify inclusion of their accounts.

The technique has real limits, and this is the misconception to correct: proof of reserves shows assets at a snapshot in time, but without a full audit it cannot prove the absence of hidden debts, borrowed coins returned after the snapshot, or omitted liabilities. It is best understood as a meaningful transparency floor rather than a solvency guarantee, and stronger schemes using zero-knowledge proofs aim to close some of these gaps.