How the Bitcoin halving works, and why supply stops at 21 million

The Bitcoin halving is a rule written into Bitcoin's code that cuts the reward paid to miners in half roughly every four years. It slows the pace at which new coins enter circulation. That slowdown is the reason the total number of bitcoin can never pass 21 million. The whole thing runs automatically, with no bank or committee in charge.

Bitcoin's most recent halving landed on April 20, 2024. Until then, miners earned 6.25 bitcoin for each block they added to the chain. After that day the reward dropped to 3.125 bitcoin, where it sits today. The next cut is expected around April 2028.

How the halving actually works

On the Bitcoin network, computers gather pending transactions into blocks. Miners race to solve a hard math puzzle for the right to add the next block, a process called proof of work. The puzzle has no shortcut. Miners simply try vast numbers of guesses until one fits, which is what makes rewriting old blocks so costly. The winner collects a set number of brand-new coins, plus the fees attached to the transactions inside that block. That fixed number of new coins is the block subsidy. A halving is the moment the subsidy gets cut in half.

Each halving follows one plain rule. Every 210,000 blocks, the subsidy drops by 50 percent. Bitcoin aims to produce one block about every ten minutes, so 210,000 blocks take close to four years to mine. Bitcoin's protocol documentation says the reward "is set to decrease geometrically, with a 50% reduction every 210,000 blocks."

No committee decides this. No vote. The schedule was fixed when Satoshi Nakamoto, Bitcoin's pseudonymous creator, launched the network in January 2009, and every full node on the network enforces it. The software most miners and users run, Bitcoin Core, simply rejects any block that tries to pay out more than the rules allow.

The halving schedule, from 50 coins to 3.125

Rewards started at 50 bitcoin per block. That number has been cut four times.

Bitcoin's first halving came on November 28, 2012, at block 210,000, cutting the reward from 50 to 25 bitcoin. The second followed on July 9, 2016, at block 420,000, dropping it to 12.5. May 11, 2020 brought block 630,000 and a reward of 6.25. The fourth halving struck on April 20, 2024, at block 840,000, and set the reward at 3.125 bitcoin.

Each step is permanent. The reward never climbs back up. Blockstream, a Bitcoin infrastructure company, puts the current pace at about 450 new bitcoin a day, down from roughly 900 before the 2024 cut. That daily flow will shrink again at the next halving.

Block 1,050,000 brings the next cut, most likely around April 2028. At that point the reward falls to 1.5625 bitcoin, and the stream of new coins halves one more time.

Why no one can name the exact date

Bitcoin does not count time in days. It counts blocks. A halving fires at a set block height, so the calendar date is always an estimate until the network gets close.

To keep blocks landing about ten minutes apart, the network tunes how hard the mining puzzle is. When more computing power joins, blocks arrive faster, and the difficulty rises to slow them down. When power drops off, the puzzle gets easier. That feedback holds the long-run average near ten minutes, but short stretches still run fast or slow. Those small gaps add up over four years, which is why a halving can arrive a couple of weeks either side of the best guess.

How to follow the next halving yourself

You do not need special software to watch the countdown. Any public block explorer shows the current block height, updated as each new block is found. Take that height away from 1,050,000, the block where the next halving lands, and the gap tells you how many blocks are left.

From there a rough date is simple arithmetic. Blocks aim to arrive about every ten minutes, so the blocks remaining translate into a stretch of time that shrinks day by day. The closer the network gets, the tighter the estimate becomes. On the day itself the reward changes the instant that block is mined, with no announcement and no switch to flip.

Why the supply stops at 21 million

Here is the part that surprises people. Nobody wrote a line of code that says stop at 21 million. The cap falls out of the halving math on its own.

Add up every reward across every four-year era and the total settles just below 21 million coins. The first run of 210,000 blocks created 10.5 million bitcoin. The next era added half that amount, then the one after added half again, and the pattern keeps shrinking. Because each era is exactly half the size of the one before, the amounts form a series that adds up to a single, finite number. Each era hands out less than the last, so the running total climbs toward a ceiling it never quite touches.

Bitcoin's protocol documentation spells out the exact figure. Because the network counts in tiny whole units and rounds down at each step, the true maximum works out to 20,999,999.9769 bitcoin. That is a sliver under 21 million. Close enough that almost everyone just says 21 million and moves on.

New coins keep trickling out for a long time yet. The rewards shrink and shrink until, by current estimates, the last fraction of a bitcoin is mined somewhere around the year 2140. The end comes gradually, not as a sudden cliff. After that, miners earn nothing but transaction fees.

A money supply no one can print

Most money works the other way around. A central bank can create more of its currency when it chooses, and that power is a normal part of running an economy. Bitcoin takes the decision away. The issuance schedule is set in software, it only ever slows, and no government or company can speed it up.

That is the case its supporters make most often. A fixed supply, laid out in advance, that anyone can check. Whether this makes bitcoin a sound store of value is a separate argument, and the answer leans on demand holding up for years. The halving guarantees the scarcity. It guarantees nothing about the price.

People often reach for gold as a comparison. Gold is scarce because it is hard to dig out of the ground, yet no one knows how much is left or how quickly new supply will arrive. Bitcoin flips that around. The coins still to come, and the exact pace they arrive at, are public and the same for everyone. That openness, more than the cap itself, is what its backers tend to stress.

What the halving means for miners

Miners feel a halving first. Their main income is the block subsidy, and overnight it falls by half. The power bill does not.

Some miners run older machines that only earn money while the reward is high. When the subsidy drops, those rigs can turn unprofitable, and their operators either upgrade the hardware or switch it off. In past cycles some firms that could not cover their costs sold coins or shut down, a shake-out traders sometimes call miner capitulation. Hashrate, the total computing power protecting the network, can wobble for weeks after a halving before it settles into a new balance.

Over time, the plan is for transaction fees to fill the gap left by the shrinking subsidy. For now the subsidy still dwarfs the fees on most days, so this is a question about the distant future rather than the next few years. Whether fees alone can pay for Bitcoin's security decades from now is an open question, and serious people in the industry disagree about it. No one has a firm answer yet, because that era is still far off.

What the halving does not do

Halvings attract a lot of hype, so it helps to be clear about the limits.

A halving does not change how Bitcoin works for everyday users. Your coins, your wallet, and your payments behave exactly as before. The one thing that shifts is the rate at which new supply arrives.

It does not promise a price move either. The reward cut is known years ahead, and markets tend to price in events they can see coming. Earlier halvings were followed by big rallies, but each one ran alongside other forces, from interest rates to the arrival of spot Bitcoin ETFs that handed large investors an easy way in. Treating the next halving as a sure payday ignores how often that bet has gone wrong.

Supply is only one half of a price. Demand is the other half, and no line of code controls that.