Skip to content

Note · Bitcoin and crypto

What is the Bitcoin halving? Why the reward keeps shrinking

The Bitcoin halving cuts in half the new bitcoin paid to miners for each block, every 210,000 blocks or about four years. It is how the supply is capped near 21 million.

This note and the comments marked the Graphite mark are Graphite’s: an AI model, Claude Opus 5.5, wrote them on 4 November 2026, following Graphite’s own instructions, from the sources at the note’s end, and checked them against those sources. Check them as you would anyone’s.

This note and its comments are written by an AI model.

Contents · 6 sections

The Bitcoin halving is the moment, every 210,000 blocks, when the number of new bitcoin paid to miners for each block is cut in half. Blocks arrive about every ten minutes, so it happens roughly every four years. It is written into Bitcoin's software, it needs no one's decision, and it is what limits the total supply to just under 21 million coins. The last halving was in April 2024, which cut the reward to 3.125 bitcoin.

In short

  • Miners are paid newly created bitcoin for each block, and that payment halves every 210,000 blocks, about every four years.
  • The halvings add up to a total of just under 21 million bitcoin, the last fraction of which should be mined around 2140.
  • Each halving cuts miners' income from new coins overnight, so over time they must rely more on transaction fees.

How does the halving work?

When a miner adds a block to Bitcoin's ledger, the block includes one special transaction that pays the miner newly created coins, called the block subsidy, plus the fees from the payments in the block. The process is described in the note on how mining works. The subsidy started at 50 bitcoin.

The rule that shrinks it is a few lines in Bitcoin Core's code. The software divides the block's height, its number in the chain, by 210,000 to count how many halvings have happened, and halves the 50-bitcoin starting amount that many times. Every node runs the same calculation. A block that pays its miner even one unit more than the rule allows is rejected by every other node, whoever mined it.

Nobody schedules a halving by date. It happens at a block number, so the date depends on how quickly blocks come. Blocks have usually arrived a little faster than every ten minutes, because computing power kept joining the network faster than the difficulty adjusted, so the halvings have come slightly under four years apart.

When have the halvings happened?

Halving Block Date reached Reward per block after Bitcoin created a day after, at 144 blocks
Start 0 3 January 2009 50 7,200
First 210,000 28 November 2012 25 3,600
Second 420,000 9 July 2016 12.5 1,800
Third 630,000 11 May 2020 6.25 900
Fourth 840,000 20 April 2024 (UTC) 3.125 450
Fifth 1,050,000 Expected around 2028 1.5625 225

The dates come from the Bitcoin Wiki's supply table. The fifth halving's date is an estimate that will firm up as the block nears.

Why does halving cap the supply at 21 million?

Because the halvings form a series that converges. The first 210,000 blocks pay 50 bitcoin each, which is 10.5 million. The next 210,000 pay 25 each, which is 5.25 million. Then 2.625 million, then 1.3125 million, and so on. Each era adds half as much as the one before, and 10.5 + 5.25 + 2.625 + … adds up to 21 million, without ever quite reaching it.

The real total is slightly less, 20,999,999.9769 bitcoin, because the software counts in whole satoshis, units of one hundred-millionth of a bitcoin, and rounds each halved reward down. After the 32nd halving the reward is a single satoshi, and the 33rd takes it to zero, at block 6,930,000, around the year 2140.

What the halvings do to new supply, worked through

The cap matters less day to day than the falling rate of new coins. Take the reward in each era, multiply by about 144 blocks a day and 365 days, and compare it with the coins already mined when that era began:

Era Reward New bitcoin a year Bitcoin mined before the era New supply as a share
2012–2016 25 about 1.31 million 10.5 million about 12.5% a year
2016–2020 12.5 about 657,000 15.75 million about 4.2%
2020–2024 6.25 about 328,500 18.375 million about 1.8%
2024–2028 3.125 about 164,250 19.6875 million about 0.8%
After 2028 1.5625 about 82,125 20.34 million about 0.4%

By the 2024 halving, more than 93% of all the bitcoin that will ever exist had already been mined. The yearly rate of new supply is now below the inflation rates most central banks aim for, and it keeps halving. These figures ignore coins that are lost for good, which make the circulating supply somewhat smaller.

Why did Satoshi design it this way?

Bitcoin needed a way to put new coins into circulation without a central bank deciding who gets them. Paying them to miners did two jobs at once: it distributed the money and it paid people to secure the network. In the white paper, Satoshi Nakamoto compared it to gold miners spending resources to add gold to circulation, and wrote that once a set number of coins had entered circulation, the incentive could move entirely to transaction fees.

When Nakamoto released the first version of the software on 8 January 2009, the announcement said: "Total circulation will be 21,000,000 coins," with the amount distributed to miners "cut in half every 4 years." That schedule has run unchanged since.

What does a halving mean for miners?

Overnight, each block pays half as much new bitcoin, while the electricity and machines cost the same. A miner whose costs were covered at 6.25 bitcoin a block has to cover them at 3.125. Unless the price of bitcoin or the fees in each block rise, the least efficient machines stop paying for themselves and are switched off, and the network's difficulty adjusts down until blocks come every ten minutes again.

That makes fees the long-run question. Each halving shrinks the subsidy further toward nothing, leaving fees to do the job the white paper foresaw for them. The design assumes that by then, people will pay enough in fees to keep miners securing the chain.

Still argued

Two questions about halvings are argued, not settled. First, whether they drive bitcoin's price: past halvings were followed by large rises, and some models built on scarcity predicted more, but a few events over a short history can't separate the halving from everything else going on, and markets know the dates in advance. Second, whether fees alone will one day pay enough to keep the network secure. This note explains the mechanism; it makes no price prediction and isn't financial advice.

Check yourself

  • Without looking back, explain why halving the reward every 210,000 blocks means the total can never pass 21 million.
  • Why do halvings happen at block numbers rather than dates, and why have they come slightly early?
  • What has to happen to fees for miners to keep securing Bitcoin once the subsidy is tiny?

Sources

Published by Graphite on , set the way Graphite publishes a note.

This note and the comments marked the Graphite mark are Graphite’s: an AI model, Claude Opus 5.5, wrote them on 4 November 2026, following Graphite’s own instructions, from the sources at the note’s end, and checked them against those sources. Check them as you would anyone’s.

About the author

Graphite

An AI model, Claude Opus 5.5, wrote it for Graphite from the sources at its end.

More in Bitcoin and cryptoSearch the MarketplaceWho writes here