Read 2026-08-31 · four exchange APIs · rebuilt daily

An Estimated 3.7 Million Bitcoin Are Lost Forever

Permanently inaccessible Bitcoin reduces actual supply below 17.3 million, lowering liquid market float while market caps still assume 21 million.

An estimated 3.7 million Bitcoin cannot be spent by any network participant. The protocol tracks unspent transaction outputs without recording key retention, meaning a permanently lost private key leaves an identical on-chain footprint to an untouched long-term wallet.

On-Chain Ambiguity and Spent Outputs

The Bitcoin ledger operates on an unspent transaction output model. Every output remains valid for spending until a cryptographic signature satisfying its locking script is broadcast in a valid block.

The blockchain cannot distinguish between three distinct physical states:

1. A private key stored in an active cold storage vault.
2. A private key destroyed in a hardware failure.
3. An output locked by an unspendable script construction.

Because the consensus layer only checks script validity at the moment of a spend attempt, inactive coins retain valid status indefinitely. Chain analysis firms estimate lost coins by measuring dormancy thresholds, but these figures remain estimates. A transaction originating from a 2010 address spent today changes the estimated lost supply retroactively without any protocol-level state change occurring prior to the spend.

Documented Permanent Loss Cases

While aggregate lost supply is an estimate, specific allocations are verifiably unspendable or documented as destroyed.

Genesis Block Coinbase Reward

On January 3, 2009, the genesis block created 50 BTC allocated to address 1A1zP1eP5QGefi2DMPTfTL5SLmv7DivfNa. The original Satoshi Nakamoto codebase did not add the genesis block's 50 BTC coinbase transaction to the global UTXO database. As a result, those 50 BTC cannot be spent by any signature.

Hardware Loss: Stefan Thomas

In 2011, developer Stefan Thomas lost the password to an IronKey encrypted hard drive containing 7,000 BTC. The drive architecture hardware-encrypts data and permanently wipes its internal cryptographic keys after ten failed password attempts. Eight incorrect attempts have been logged, leaving 7,000 BTC inaccessible.

Physical Loss: James Howells

In mid-2013, IT worker James Howells accidentally discarded a laptop hard drive containing private keys for 7,500 BTC mined in 2009. The drive was buried under municipal landfill waste in Newport, Wales. Without the drive's physical access, the corresponding address outputs cannot generate valid signatures.

Proof-of-Burn Addresses

Outputs sent to deliberately invalid public keys, such as 1CounterpartyXXXXXUWLpVZacawc1, contain no matching private key. The Counterparty protocol burned 2,130.86 BTC in January 2014 by sending funds to this address to issue protocol tokens. These coins are mathematically unspendable.

Supply Math and Effective Market Float

Calculations for Bitcoin market capitalization routinely use total mined supply, currently approaching the 21,000,000 BTC theoretical limit. Subtracting estimated lost supply changes the denominator for liquid valuation models.

• Theoretical maximum supply: 21,000,000 BTC
• Estimated permanently lost supply: 3,700,000 BTC
• Effective maximum spendable supply: 17,300,000 BTC

Dividing 3,700,000 by 21,000,000 shows that 17.62% of all Bitcoin that will ever exist is permanently removed from circulation.

When traders evaluate spot liquidity relative to derivative open interest, using unadjusted supply overstates available market float. An illiquid supply base means order book depth on spot exchanges is concentrated across a smaller real circulating token pool than public market capitalization figures suggest.

Derivatives Funding Rates and Capital Efficiency

Because real spot supply is tighter than nominal token supply implies, perpetual swap funding rates reflect localized supply imbalances across venues.

Below are the live normalized 8-hour funding rates and published fee schedules across major venues:

Carrying Cost Worked Example

Holding a long position in BTC perps requires accounting for both funding rate spreads and venue execution costs.

Consider a $100,000 BTC long position held for 30 days (90 funding intervals of 8 hours each):

• OKX BTC funding rate: +0.0078% per 8h
• Bitget BTC funding rate: +0.0100% per 8h
• Spread: 0.0022 percentage points per 8h

Math for 90 intervals on OKX: 100,000 multiplied by 0.000078 equals $7.80 per interval. 7.80 multiplied by 90 equals $702.00 total funding cost.

Math for 90 intervals on Bitget: 100,000 multiplied by 0.000100 equals $10.00 per interval. 10.00 multiplied by 90 equals $900.00 total funding cost.

The spread between OKX and Bitget costs long position holders $198.00 per $100,000 of position size over 30 days.

Execution fees add to entry and exit friction. Opening and closing a $100,000 position on OKX as a taker costs 0.0500% per side, totaling 0.1000% or $100.00. Opening and closing the same position on MEXC as a taker costs 0.0200% per side, totaling 0.0400% or $40.00.

Impact on Liquidation Risk

A reduced real circulating float increases the market impact of large market orders. If 17.62% of total supply is permanently unspendable, price slippage during cascading liquidation events accelerates faster than linear order book depth models predict.

For a trader utilizing 20x leverage, a price move of 4.5% against the entry position triggers maintenance margin liquidation. In markets with artificially high market cap metrics masking a restricted liquid float, sudden spot order book imbalances clear out bid depth faster, driving mark prices directly into liquidation triggers before derivative arbitrage can rebalance the venue spread.

Get 20% fee rebate on Bitget →20% of your trading fees back, on every product. The rebate comes out of the commission I would otherwise receive, so it costs you nothing. Affiliate link — see the footer.