What Is Proof of Reserves and Can It Prove an Exchange Is Solvent?

Proof of reserves is a cryptographic check that lets a crypto exchange prove it actually holds the coins its customers are owed — for every 1 BTC credited to user accounts, it shows at least 1 BTC sitting in wallets it controls. It proves the assets exist. It does not prove the exchange is solvent. That gap is the whole story.

After the FTX collapse in November 2022, "proof of reserves" became the transparency badge every major platform rushed to wear. But a reserves proof answers only half the solvency question, and the half it skips is the half that sinks exchanges. If you keep coins on any platform, understanding what this proof really covers is part of basic self-defence — the kind of foundation a structured cryptocurrency and crypto trading course is built to give you instead of scattered threads.

What does proof of reserves actually prove?

Strip away the branding and a proof of reserves is a statement about one side of a balance sheet: the asset side. The exchange is saying, "here are wallets I control, and the coins in them are at least as large as everything my customers are owed." When done well, it is verifiable by anyone, not just a claim in a blog post.

Two things have to be shown together for the proof to mean anything. First, the exchange has to prove control of the reserve wallets — usually by signing a specific message with the private keys of those wallets, which only the true owner could do. Second, it has to prove the total it owes customers, the liability figure those reserves are being compared against.

That second number is where the clever cryptography lives. An exchange cannot simply publish every customer's balance — that would be a privacy disaster. So it compresses all customer balances into a single fingerprint called a Merkle root, which can be checked by each user without exposing anyone else's holdings. We will walk through exactly how that compression works in the next section.

Here is the subtle part most beginners miss: proving assets exceed a published liability number is only honest if the published liability number is complete. Nothing in a basic proof of reserves forces an exchange to include every debt it owes. It can prove it holds a lot of Bitcoin while quietly leaving money it borrowed, or losses it has hidden, off the list entirely. The proof will still verify. That is the FTX-shaped hole we will keep returning to.

How does a Merkle-tree proof of reserves work?

A Merkle tree is a way of boiling thousands or millions of numbers down to one short code, in a way that lets any single number be checked against that code. The proof-of-reserves process built on it runs in four stages.

Stage one: hashing. Each customer's account ID and balance are run through a one-way hash, usually with a random value added so nobody can reverse the code back into a real balance. This is the leaf at the bottom of the tree.

Stage two: building the tree. Those leaf codes are paired and hashed together, then the results are paired and hashed again, level after level, until a single code remains at the top. That final code is the Merkle root, and it mathematically depends on every balance beneath it.

Stage three: publishing. The exchange publishes the root together with proof it controls wallets holding at least that much. Because the root changes completely if even one balance is altered, the exchange cannot quietly shrink its stated liabilities after the fact without everyone noticing.

Stage four: user verification. You can log in, pull the short path of codes linking your own balance to the published root, and confirm your account was counted. If millions of users each check their own leaf, the total liability figure becomes very hard to understate. This self-custody mindset is the same reason we stress guarding your own seed phrase — verification only helps the people who actually do it. The cryptography here is a cousin of the hashing that secures blockchains themselves, which we unpacked in our explainer on proof of stake versus proof of work.

Can proof of reserves prove a crypto exchange is solvent?

No — and this is the single most important thing to understand. Solvency means assets are greater than liabilities. A standard proof of reserves rigorously shows the assets and only loosely shows the liabilities, so it can be completely accurate and still leave you blind to a failing exchange. Three weaknesses do the damage.

Source: Cointelegraph and Banxa analyses of proof-of-reserves limitations, 2025.

Weakness one: it proves assets, not liabilities. You cannot prove a negative. An exchange can show a billion dollars of Bitcoin on-chain while owing three billion in obligations that never appear in the proof. The assets are real; the solvency is fiction.

Weakness two: it is a point-in-time snapshot. The proof photographs one instant. An exchange short of funds can borrow coins just before the snapshot and return them the next day. The proof is technically accurate for that frozen moment and completely misleading about every other moment.

Weakness three: it cannot see borrowed or encumbered coins. Signing a wallet message shows access right now, not that the coins are unpledged and genuinely the exchange's own. The same borrowed stack could even be shuffled between related companies. FTX's own reserve claims looked reassuring precisely because they left the ugly liabilities off the page. A proof is only as honest as the liability list behind it — which is why a sudden loss of confidence can drain an exchange faster than any report updates, the same reflexive panic that drives a stablecoin to lose its dollar peg.

Which Indian crypto exchanges publish proof of reserves?

Adoption in India is real but still patchy, and you should treat every figure below as dated and attached to a specific report rather than a permanent fact. CoinDCX was the first Indian exchange to publish a proof of reserves and has moved to a regular cadence — quarterly reserve disclosures alongside monthly transparency reports. Its January 2025 disclosure stated roughly 1.48% in excess reserves above customer liabilities.

CoinSwitch has also published reserve reports, making it one of the few Indian platforms doing so at the time of writing. Across the wider set of Indian exchanges, though, published proofs remain the exception rather than the rule, so the absence of one is not proof of trouble, and the presence of one is not proof of safety.

WazirX is the cautionary tale closest to home. It had published reserve information earlier, with a large share of user assets reported as held through Binance wallets. Then a hack in July 2024 drained roughly 230 million dollars, froze withdrawals for months, and forced a restructuring before the platform relaunched in October 2025. No reserves snapshot taken before that breach would have warned an ordinary user about the custody and security risk that actually mattered.

If you are weighing where to keep coins at all, our comparison of centralised versus decentralised exchanges in India sets out the broader trade-off a proof of reserves can never settle on its own.

How do you read a proof-of-reserves report without fooling yourself?

Treat the report as one input, not a seal of approval. A few practical habits separate a careful reader from someone comforted by a logo:

• Check who signed off. A self-published root is weaker than one attested by an independent third-party auditor who also examined the liability side. Reserves plus an audited liabilities view is far closer to real solvency assurance.
• Look at frequency and dates. One proof from eight months ago tells you about a moment long gone. A quarterly or monthly cadence is harder to game with a single borrow-and-return trick.
• Verify your own balance. The design only works if users actually check their leaf against the published root. If you never verify, you are trusting, not checking.
• Ask what is not shown. Does the report address liabilities, borrowings and customer-fund segregation, or only the shiny asset total? Silence on liabilities is the loudest signal.
• Remember the snapshot limit. Even a perfect proof says nothing about what happens the hour after it is taken. Position sizing and not over-concentrating on one platform matter more than any single report.

None of this requires you to be a cryptographer. It requires you to know which question each piece of evidence answers — and which questions nobody has answered yet.

What to do next

Proof of reserves is a genuine step forward from the pre-FTX era, when exchanges asked for blind trust. Used well, with third-party attestation and a regular cadence, it meaningfully raises the cost of lying about assets. But it is a floor, not a guarantee, and it was never designed to certify solvency. The careful user reads it for what it proves, discounts it for what it misses, and keeps the self-custody option firmly in view.

If you want to move from headline-level understanding to actually evaluating platforms, wallets and on-chain evidence yourself, a structured path beats piecing it together from scattered posts. That is exactly what NIFM's cryptocurrency training is built to deliver.

Frequently Asked Questions

What is the difference between proof of reserves and a full audit?

Proof of reserves cryptographically shows the assets an exchange controls and, in the stronger versions, compares them to customer liabilities. A full financial audit goes further — examining all liabilities, borrowings, internal controls and related-party dealings. Proof of reserves is faster and more frequent; a full audit is deeper but rarer. The most credible platforms combine both.

Why can proof of reserves not prove solvency on its own?

Solvency means assets exceed all liabilities. A basic proof of reserves verifies assets strictly but relies on the exchange to disclose its liabilities honestly and completely. Because it cannot force a full, truthful liability list and only captures a single moment, an exchange can pass the check while still being deeply insolvent, exactly the FTX pattern.

How can I verify my own balance in a proof of reserves?

Many exchanges give logged-in users a verification tool that shows the short chain of hashes linking your account balance to the published Merkle root. You confirm your balance was included in the total the exchange claims to back. If every user did this, understating liabilities would be far harder. The system only protects people who actually check.

Do Indian crypto exchanges have to publish proof of reserves?

There is no blanket legal requirement in India forcing every exchange to publish a proof of reserves at the time of writing. Some platforms, such as CoinDCX and CoinSwitch, publish them voluntarily as a trust signal, while many do not. Registration with financial-intelligence authorities and tax rules are separate obligations from voluntary reserve disclosure.

Is my crypto safe if an exchange shows proof of reserves?

Safer in one narrow sense — you have evidence the assets existed at the snapshot — but not fully safe. It says nothing about hidden debts, future moments, hacks or how well customer funds are segregated. For long-term holdings, many users move coins to self-custody rather than rely on any exchange report, however reassuring it looks.

Disclaimer: This article is for educational purposes only and does not constitute investment advice. Markets carry risk — please do your own research or consult a qualified financial professional before investing. NIFM provides training and exam preparation; certification exams conducted by regulatory or professional bodies are administered by those bodies independently.