What Is Proof of Reserves? What It Shows and What It Does Not

A proof of reserves has two halves, and both are needed:

1. Assets. The exchange shows it controls certain wallets on the blockchain, usually by signing a message from them or by moving a small test amount. Anyone can then look up the balances on the public ledger.
2. Liabilities. The exchange lists what it owes every customer. To keep balances private it puts them into a Merkle tree, a structure that lets you check that your own balance is included without seeing anyone else's. The tree's single top value, the root, is published.

Then compare: assets should be at least equal to liabilities. A report with only the assets half proves nothing about what is owed.

Only about a third of the exchanges publish a proof we could confirm. Of those, a little over half include liabilities, and few name who checked it. The wider check was a limited search with no score, so its "not verified" count is higher. Dates count only where we found one. Details exchange by exchange: proof of reserves compared.

1. Are liabilities included? Assets alone do not show solvency.
2. Which assets are covered? Many reports cover only the major coins, not everything you might hold.
3. Who checked it? An independent firm is better than the exchange alone. Look for the name.
4. How recent is it? A report from two years ago says little about today.
5. Can you verify your own balance? A proof you can check yourself is stronger than a figure you must trust.

• It is a snapshot. It says nothing about the days between snapshots.
• It is usually an "agreed-upon procedure" or attestation, not a full financial audit.
• It may leave out debts that are not customer balances, such as loans.
• It does not measure security, regulation or how the exchange handled past incidents. Those sit next to it in the Exchange Rigor Report.