Key takeaways
Bitcoin’s 21-million cap reflects issuance rules, not how much BTC is actually usable or spendable.
A meaningful share of supply is unavailable due to unmined coins, provably unspendable outputs and lost private keys.
Major researchers estimate millions of BTC are effectively removed from circulation, even under conservative assumptions.
The economic reality of Bitcoin’s supply is best understood as the headline cap minus future issuance, unspendable coins and lost or inaccessible BTC.
“21 million” is Bitcoin’s headline cap, a rule baked into the protocol’s issuance schedule through a block subsidy that halves every 210,000 blocks. But headline supply is not the same as usable or spendable supply.
Once you account for coins that (1) have not been mined yet, (2) are provably unspendable, and (3) are effectively inaccessible due to lost keys or catastrophic custody failures, Bitcoin’s “real” supply is meaningfully lower than the number most dashboards display.
21 million is an asymptote, and the protocol may never reach it exactly
Bitcoin’s issuance comes from the block subsidy function (or block reward), which started at 50 Bitcoin (BTC) and is cut in half every 210,000 blocks. That schedule is implemented in Bitcoin Core and is the mechanical reason people refer to a 21 million cap.
Two practical wrinkles matter here:
Timing: Halvings occur every 210,000 blocks, not every four years by calendar, so the exact dates drift with block production.
Rounding and integer math: Because the subsidy is denominated in satoshis and repeatedly halved using integer arithmetic until it truncates to zero, the final issued amount is slightly below 21,000,000 BTC. It is often cited at approximately 20,999,999.9769 BTC in technical explanations.
So, even before accounting for “lost coins,” the system’s maximum issued supply is best understood as “about 21 million,” not a guaranteed exact integer.
A big chunk of Bitcoin has not yet been mined
Bitcoin’s circulating supply is typically defined as the number of coins mined to date, and that figure remains below the cap. For example, CoinMarketCap currently lists Bitcoin’s circulating and total supply at about 19.96 million BTC, with 21 million shown as the maximum supply.
Blockchain.com’s Total Circulating Bitcoin chart similarly tracks how much BTC has been issued so far, highlighting that “in circulation” usually means “created,” not “accessible.”
This is the most straightforward reason the “real” supply is lower than 21 million: A non-trivial remainder is still scheduled to be minted through future halvings, stretching toward the year 2140 in many standard explanations of the issuance curve.
Some BTC are provably unspendable
Not all issued coins are even theoretically spendable.
The genesis block’s subsidy is unspendable. Bitcoin’s first block contains a subsidy output that cannot be spent under Bitcoin’s rules and implementation quirks, and it is widely treated as permanently unspendable.
OP_RETURN outputs can intentionally create provably unspendable outputs. Bitcoin’s developer documentation describes “null data” outputs using OP_RETURN as unspendable, meaning they cannot be spent. This is a standardized way to mark outputs as provably unspendable and effectively burn any satoshis assigned to them.
An important nuance for creators and builders: Not every “burn” on Bitcoin is equally provable. Discarding a private key can make coins practically unspendable, but it is not cryptographically obvious to the network in the way an OP_RETURN-style output is.
The largest supply cut comes from coins that are likely lost or inaccessible
Bitcoin custody is asymmetric: If you lose your keys, there is no central authority to reset them. As a result, a meaningful fraction of mined BTC appears to be permanently lost.
What counts as “lost” is difficult to prove onchain since inactivity does not equal loss. Credible estimates are therefore usually presented as ranges, derived from heuristics, early wallet behavior and documented loss events.
In 2018, Chainalysis published an estimate suggesting that between 2.3 million and 3.7 million BTC had been lost in the context of analyzing Bitcoin’s effective money supply.
River published a public estimate in 2023 stating that 3 million-4 million BTC is “irreversibly lost,” framing this as a material share of total supply.
CoinShares estimates that approximately 1.58 million BTC could be classified as lost as of March 2025.
The common thread across these sources is that even conservative lost coin estimates remove millions of BTC from the spendable supply. The economic reality is therefore best described as 21 million minus not-yet-mined coins, minus unspendable coins, minus lost or inaccessible coins.
“Circulating supply” dashboards are not misleading
CoinMarketCap defines “circulating supply” as the best approximation of assets available to the public. In practice, however, Bitcoin’s circulating supply remains anchored to coins that have been issued or mined, minus limited special-case adjustments, rather than a cryptographic proof that those coins are accessible or liquid.
Those definitions do not claim that:
All coins are liquid
All coins are accessible
All coins can be sold or moved.
As a result, two statements can be true:
19.96 million BTC exist, meaning they have been mined.
The spendable or accessible supply is materially lower due to provable burns and likely lost coins.
For investors, this means circulating supply figures should be interpreted as technical issuance data, not as a direct measure of market liquidity.
What this means for Bitcoin investors and miners
For investors: Scarcity is structural, but supply metrics require nuance
For investors, the fact that Bitcoin’s effective supply is lower than the 21 million headline cap reinforces a key structural property: Bitcoin’s scarcity is determined by accessible coins, not theoretical issuance limits.
While the protocol enforces a hard ceiling on creation, a portion of issued Bitcoin is permanently removed from economic circulation due to provable burns and lost private keys. As a result, the number of BTC that can realistically be held, transferred or sold is smaller than commonly cited supply figures.
For miners: Issuance rules remain unchanged, but accessible supply dynamics evolve
For miners, the fact that Bitcoin’s effective supply is lower than 21 million does not change block rewards, halving mechanics or consensus rules. The protocol continues to issue new Bitcoin according to the predefined subsidy schedule, regardless of how many older coins are lost.
However, a reduced effective supply still has indirect relevance:
Newly mined Bitcoin is fully usable. Block rewards are immediately spendable, making miners the primary source of new, liquid BTC entering the market.
Halvings still control supply. While lost coins reduce usable supply over time, halvings remain the main mechanism that slows new Bitcoin issuance.
Miner economics do not change. Lost Bitcoin does not increase block rewards or fees; miner revenue continues to depend on issuance, transaction fees and overall network activity.