Bitcoin depends on a vast network of mining equipment: computers able to perform complex cryptographic work very quickly, to “discover” and validate new blocks and mint new Bitcoins.
But the supply of Bitcoins is limited by the network’s source code: there will never be more than 21 million BTC. What happens when they have all been mined?
First, let’s orient ourselves technically. When we talk about BTC mining, what are we actually talking about?
How mining works on the Bitcoin network
Bitcoin is a Proof-of-Work blockchain. It’s a chain of discrete blocks of information, each of which is agreed upon by the whole network, and contains information about all previous blocks. That makes it extremely secure and extremely inefficient. Miners identify groups of transactions awaiting validation in the Bitcoin memory pool and validate them.
Validating a transaction group summing around 1 megabyte (“block”) involves work and attracts a reward, so there’s competition. It’s done by trying to match a 64-digit hexadecimal number, called a “hash”, that’s nearly equal to the target hash. The potential number of these is huge, so miners have to cycle through their guesses quickly. This requires a lot of computation, and the first miner to solve the requisite hash gets the reward for validating the block.
The reward for BTC block validation is currently 6.25 BTC per block, but there’s more to this part of the story.
BTC rewards for block validation halve about every four years, from an initial 50 BTC when the network launched. The current reward, 6.25BTC, is about $197,989, so mining remains profitable. The reward last halved in 2020.
|Year||Block Reward (in BTC)||Reimagining Unmined BTC|
The block reward halving tracks reasonably well with halvings in the remaining number of unmined BTC. Clearly this also means the number of BTC mined per year falls by about 50% every four years too. This has profound implications for the future of BTC mining because it lets us know how long it will probably take to mine all available BTC. This is forecast to be about 120 years, in 2140.
Transaction fees and miners’ income
However, the falling block reward rate could have other effects, including slowing mining further than forecast. Just over 88% of all available BTC has already been mined, meaning that an ever-bigger network (and one with ever-growing economic implications) is maintained by miners receiving ever-reduced rewards for ever-increasing work.
Miners are in a race with declining rewards as well as with each other. On their side is the falling price of computing power, both as available power per unit rises in accordance with Moore’s law and as the number of available units rises.
In addition, miners earn transaction fees for validating blocks; these fees make up an increasing portion of miners’ incomes, so that miners typically mint around 900 BTC per day at the time of writing, but earn between 60 and 100 BTC per day — slightly more than mining rewards alone could supply.
In fact, miners derive about 6.5% of their total income from transaction fees, with the remainder coming from direct mining rewards; losing mining as a source of revenue would make block validation uneconomical for many miners.
However, as mining rewards decline, transaction fees will rise. In 2010, a 0.1BTC transaction fee floor was introduced via a code change, but this was removed as transaction volume increased. Miners earn transaction fees from validating the transactions in a block, so transaction fees and block rewards go to the same miner.
Transaction fees have risen in dollar terms and fallen in BTC terms since the network’s inception, as transaction volume has risen and the dollar value of BTC has outstripped that growth.
It’s also worth noting that transaction fees are significantly less predictable than block rewards:
Sharp increases in transaction fees track transaction volume, not BTC dollar value. That’s because transaction fees are set dynamically, based on the data volume of a transaction and the congestion of the network.
But transaction fees are set in BTC, and BTC’s value has appreciated over time, despite volatility:
Earning power: electricity and consensus for Bitcoin mining
The BTC network isn’t the only source of economic pressure on Bitcoin miners. The need for low-priced, reliable electricity networks and easily-cooled physical locations for the huge accumulations of computing equipment required by industrial BTC mining has tended to make mining concentrate in just a few locations, chiefly in China.
So China’s decision had major repercussions for the BTC network. Hash rate fell by 40% as Chinese miners went dark.
Currently, though, mining is in the process of withdrawing from China following a crackdown that saw the industry effectively banned in several provinces. It’s moving instead to the US and Kazakhstan.
Miners struggle with electricity costs, which can eat into their profit margins. The average cost of a single BTC transaction in the United States is about $200. The future cost of electricity is therefore a major issue; how will Bitcoin miners cope with significantly higher electricity costs?
One answer might be in the burgeoning growth of green energy, whose cost per unit promises to be much lower. In addition, the energy efficiency of BTC mining equipment is improving all the time, meaning electricity costs might recede as a concern for miners long before the final Bitcoin is mined.
In fact, even as the Bitcoin network has hoovered up increasing quantities of electricity, the cost to mine one BTC has never increased in real terms.
Another possibility lies in Bitcoin’s now-archaic PoW consensus algorithm. Since Bitcoin emerged in 2009, Proof-of-Work has been superseded by more energy- and computation-efficient algorithms for blockchains, such as Proof-of-Stake; longtime number-two digital asset and groundbreaking general purpose blockchain Ethereum is planning to switch to PoS, and newly-launched projects like Solana and Polkadot prefer it too.
So could Bitcoin move to PoS? There’s quite a fierce debate about whether it’s a good idea, but it’s technically possible. However, it’s not a requirement for Bitcoin to remain profitable.
The inflection point: are we there already?
The final Bitcoin might be forecast to be mined in 2140, but a lot can change in 120 years. Realistically, we’re less concerned about what happens when all 21 million BTC are mined, and more concerned about some potential transition point encountered en route.
This chart shows the supply of BTC dwindling and block rewards halving in a series of ever-shallower steps. But it also shows, more clearly than I could say, the shape of BTC supply. Long before all the BTC runs out, there will be a long period of minimal mining. We’ll pass 20 million somewhere around 2025. The remaining million Bitcoins will take nearly a century to mine.
Analyses of remaining BTC reserves also tend to ignore the relatively large number of BTC which have been irrevocably lost already. Whether to people who died without disclosing their private keys, or to lost private keys, or other reasons, estimates of permanently-lost BTC ran as high as nearly 4 million, as early as 2017.
Current estimates indicate that just 14 million BTC will ever circulate, because of the large number of losses. So we may already be close to a situation of supply equilibrium, where we’re losing as many BTC as we’re mining. In other words, we may be undergoing the supply inflection now, and the mining inflection will be within a decade or two rather than a century.
Outlook: what will happen to validation on Bitcoin?
When that happens, what will miners do? Let’s assume no fundamental changes to the BTC network. Probably, miners will continue to validate blocks, chasing transaction fees. Smaller miners will either specialize early or die off, but mining is already an industrialized concern with huge conglomerates. They will pivot, moving to focus on transaction fees. These are dependent on volumes and congestion, so they’ll fluctuate. The risk here is the natural tendency for specialized plant to concentrate; it’s possible that a 51% attack could become more likely as BTC validators, no longer primarily miners, consolidate.
In the meantime, mining continues to be profitable. Miners are paid in BTC, so a miner in 2016, rewarded with 12.5 BTC for validating a single block, made $5,425. A miner who validates a block today will be paid “half” that: 6.25 BTC, or $197,989. That’s some decline.
And as miners wean themselves from validation rewards to transaction fees, the network may simply respond to accommodate the change. When China’s miners went dark and the network hashrate nosedived, something interesting happened too: transaction fees rose. The price of transactions rose to its highest ever level, an average of $59 per transaction, because of increased congestion on the network. This suggests a mechanism whereby mining activity may simply fall until it’s profitable again.
Will that be a big issue for BTC? Probably not. After all, BTC is increasingly a low-transaction network; an exchange for an investment asset, not a tool for making purchases. That tendency may simply be exacerbated as mining activity on the network falls and transactions become more expensive. And we’ve seen price setting before; that could happen again in the future, if it’s required to keep the network functional.