Bitcoin: is the 21 million cap threatened?

In December 2024, asset manager BlackRock released a three-minute video explaining how bitcoin works. While the video wasn’t newsworthy, it included a controversial disclaimer- “there are no guarantees that bitcoin’s maximum supply of 21 million won’t change”. Given the hard cap’s role in bitcoin’s scarcity, BlackRock unintentionally reignited a debate among the crypto community about its immutability.

Is the Bitcoin hard cap coded?

Satoshi Nakamoto, Bitcoin’s pseudonymous founder, didn’t explicitly mention the hard cap in its source code, the instructions for how the protocol works (other key components include the consensus mechanism and the number of transactions recorded on each block). Instead, it’s determined by events the crypto community has dubbed ‘halvings’.

Miners earn block rewards in the form of bitcoin in return for their participation in the ‘proof of work’ consensus mechanism. This is the only way new coins come into circulation. These rewards halve every time 210,000 blocks are mined, which takes around four years. The final bitcoin is expected to be mined in 2140, although rounding errors mean that the total supply will be slightly less than 21 million- roughly 20,999,999.9769.

Satoshi set the initial block reward at 50 btc, meaning over 10 million btc were issued before the first halving in 2012 (210,000 blocks x 50 btc = 10.5 million btc). As the tab below shows, by the sixth halving, which should take place sometime in 2032, 99% of the supply will be in circulation. Continuing this calculation would show that the block reward will fall to a fraction of one Satoshi (100 millionth of a bitcoin) by 2140.   

Can the Bitcoin hard cap be changed?

Theoretically, yes, but it would require several things to fall into place.

Firstly, developers would have to submit a Bitcoin Improvement Proposal (BIP). Given the controversy such a change would cause, it would likely trigger heated debate among the community. The BIP would also face thorough testing and review. If approved, developers would need to integrate the changes into the Bitcoin Core, the application that puts the source code into action, used by roughly 97% of bitcoin nodes (the computers storing a copy of the blockchain).  

The next step would be to find consensus among the nodes, and bitcoin’s decentralised governance model sets a high bar for major changes like removing the hard cap. Estimates vary, but in June 2024 Bitnodes reckoned that nearly 22,000 nodes run the bitcoin software. The majority would have to upgrade their version of Bitcoin Core for the change to take effect, or they could simply reject the update.

Changing the hard cap would also leave miners with a dilemma because it requires a ‘hard fork’, a network update that creates a new chain running alongside the original. The new chain would be incompatible with the previous version, forcing miners to choose which one to support. In contrast, ‘soft forks’ are updates that don’t alter the underlying structure so the blockchain can continue to operate uninterrupted.

The 2015-17 ‘blocksize war’ is a good example of the upheaval a controversial update can cause. As bitcoin gained traction, the limited amount of data each block stored became a hindrance, making transactions slower and more expensive. Divisions emerged in the community: the grassroots supported maintaining the block size to preserve bitcoin’s security and decentralisation while corporate interests wanted to increase it so the network could scale. After what many consider bitcoin’s first civil war subsided, a hard fork created Bitcoin Cash, which used a bigger block size, and a soft fork changed the way the original chain stored data on blocks (the ‘Segregated Witness’ update). For the record, Bitcoin Cash’s market capitalisation is $6.5 billion as of February 2025 while Bitcoin is worth just under $2 trillion. 

This episode reminds that stakeholders of Bitcoin networks, individual holders included, have their word when it comes to bitcoin consensus decisions and economics disincentivise any changes to the hard cap. Increasing it would lead to inflation and damage bitcoin’s scarcity, one of the primary catalysts driving its value. While miners would most benefit from an increased supply, they would ultimately lose out because bitcoin’s price would fall in fiat terms, and they have to spend significant sums of fiat currency to operate. Research shows that miners invested $3.6 billion in property, plant and equipment in 2024, while the 175 terawatt hours of energy they consume yearly costs nearly $9 billion, based on an average (and relatively cheap) rate of $0.05 per kilowatt hour.

What about lost Bitcoins?

While nearly 20 million bitcoins have been issued, that doesn’t mean they’re actively circulating. Using wallet inactivity as a proxy, crypto platform River Financial estimates that almost 1.6 million coins (worth over $1.5 billion as of February 2025) have effectively been removed from circulation because the holders have mismanaged custody. For instance, losing your private keys, a string of random letters and numbers which is the crypto equivalent of a password, locks you out of your wallet. ‘Not your keys, not your coins’, as the saying goes. Most of the lost bitcoins sit in wallets that have been dormant for more than 10 years.

Incidentally, this figure doesn’t include Satoshi’s coins. The founder holds nearly one million bitcoins, earned from mining early blocks, in around 20,000 wallets. They haven’t moved since Satoshi apparently stopped mining in 2010, although one of the wallets received 27 bitcoins (worth just over $1.2 million at the time) in January 2024.

Based on these figures, the hard cap is closer to 18.5 million.

In one famous case, Welsh computer engineer James Howells lost a reported eight thousand bitcoins (worth almost $800 million as of February 2025) that he received as block rewards when his then partner threw away the hard drive storing his private key in 2013. Howells has spent more than a decade trying to convince Newport City Council to allow him to search the site, even offering to let it keep 25% of the value of the haul. Media reports suggest he’s considering purchasing the site after the council announced it will close sometime in the next two years.  

It’s unlikely even quantum computing, which has the potential to solve problems that the most powerful computers struggle with today, will recover lost bitcoin. Concerns have emerged that these supercomputers will be able to figure out a user’s private key from their public key, the wallet address shared with senders. However, as Christopher Bendiksen, CoinShares’ Bitcoin Research Lead, recently wrote, this threat is neither imminent nor problematic. There are several ways to tackle it, such as hiding public keys behind a hash (created by a cryptographic technique that converts the key into a fixed-length string of numbers and letters) and implementing a soft fork introducing a new address format safe from quantum computing.

Is a change of unit more likely?

Shifting a decimal point is a more realistic proposition than lifting the hard cap. John Carvalho, CEO of software developer Synonym, submitted a BIP at the end of 2024 that proposed changing bitcoin’s unit of measurement. Instead of one bitcoin equalling 100 million Satoshis, one bitcoin would equal one Satoshi. So if you spent $250, you’d receive 255,000 btc rather than 0.002550025 btc (as of February 2025).

Here's Carvalho’s motivation:

By redefining the smallest unit as "one bitcoin," this BIP aligns user perception with the protocol’s true nature. It reduces cognitive overhead, ensures users understand Bitcoin as counting discrete units, and ultimately improves educational clarity and user experience.

Earlier BIPs have made similar proposals.

Changing the denomination would help to overcome unit bias, a concept rooted in behavioural finance suggesting people prefer to purchase a whole unit of an item, regardless of its size. In crypto terms, it implies people are reluctant to buy bitcoin because of its eye watering valuation ($96,138 as of February 2025), potentially threatening adoption.

Another way to deal with unit bias is by purchasing exchange-traded products (ETPs) which track the price of bitcoin and trade on mainstream exchanges, so you don’t have to purchase directly from unregulated or underregulated crypto exchanges. ETPs are sold in shares, so buying one for, say, $30 might feel more satisfying than receiving 0.000304632 btc (as of February 2025). Getting rid of the integers would also make transactions easier to understand.

In summary: a change of unit more likely than a change of supply

So to sum up, while bitcoin’s hard cap could theoretically increase, governance challenges allied with incentives put in place by Satoshi when designing the protocol suggest it’s unlikely to happen. Even if the concentration of coins in the hands of a few ETF issuers or corporations could undermine the consensus around the hard cap, the underlying owners of their shares could simply sell them to express a different opinion and choose the legitimate chain. If anything, the circulating supply will be even lower, given the number of coins lost to self-custody mismanagement and the size of Satoshi’s stash. But there’s an argument in favour of introducing a new unit of measurement to avoid unit bias and overcome bitcoin’s perceived limited supply.