Course outline · 10 lessons
Bitcoin Mining & Economics

Bitcoin Halving Explained: The Supply Schedule That Makes Bitcoin Scarce

Four unmarked aluminium bars with successively halved lengths lie in parallel on a grey stone surface.

The Bitcoin halving is the rule at the center of Bitcoin’s monetary policy. Every 210,000 blocks, roughly every four years, the number of new bitcoins a miner receives for adding a block is cut in half. This shrinking flow of new coins is what keeps the total from ever exceeding 21 million bitcoin. If you’ve read how Bitcoin mining works, the halving completes the picture: it is the rule that sets how much new bitcoin mining pays, block after block, for the next century.

What Is the Bitcoin Halving?

When Satoshi Nakamoto launched Bitcoin in January 2009, a miner earned 50 BTC for each block. That generous reward helped start the network when bitcoin had no market price. But the software also included a rule: after every 210,000 blocks, the new-coin reward (the block subsidy) is cut by half.

There is no committee and no vote behind this. Every full node checks every block it receives, and a block whose miner pays itself more than the current subsidy plus fees is rejected as invalid, however much work went into it. To let miners create more coins than the schedule allows, people would have to release software with different rules and persuade users, exchanges, businesses and miners to run it. Anyone who kept the old software would treat the extra coins as invalid, and the network would split in two. That is called a hard fork, and each user decides for themselves which rules they follow. (Rules can also be tightened without a split, which is how upgrades such as Taproot in 2021 were made; who enforces Bitcoin’s rules explains the difference.)

Two blocks reach the full nodes. The rule since April 2024 is that a block may pay its miner 3.125 BTC plus fees. The first block pays its miner 3.125 BTC plus fees: every node accepts it and adds it to its copy of the chain. The second pays its miner 6.25 BTC plus fees: every node rejects it, however much work went into it, and the miner gets nothing. There is no committee and no vote: each node checks every block for itself.
The schedule is enforced by every node, not by a vote. A block that pays its miner more than the subsidy plus fees is invalid, however much work went into it. Each node rejects it on its own.

The schedule so far:

  • Blocks 0–209,999: 50 BTC per block
  • Blocks 210,000–419,999: 25 BTC per block
  • Blocks 420,000–629,999: 12.5 BTC per block
  • Blocks 630,000–839,999: 6.25 BTC per block
  • Blocks 840,000–1,049,999: 3.125 BTC per block (the current era, since April 2024)

Bitcoin aims for one block every 10 minutes on average, so 210,000 blocks take about four years. The real timing drifts a little: when hashrate grows quickly, blocks arrive slightly faster than every 10 minutes until the difficulty adjustment catches up.

A staircase of the new bitcoin in each block, the subsidy. 50 BTC from block 0 in January 2009; 25 BTC from block 210,000 in November 2012; 12.5 BTC from block 420,000 in July 2016; 6.25 BTC from block 630,000 in May 2020; 3.125 BTC from block 840,000 in April 2024, the current era: in November 2026 more than 970,000 blocks have been mined. The next halving, at block 1,050,000 around 2028, cuts it to 1.5625 BTC. Each step is 210,000 blocks, about four years, and each is half the one before.
Every 210,000 blocks, the new bitcoin in each block is cut in half. That is about every four years. Miners now get 3.125 BTC per block; from the next halving, around 2028, they will get 1.5625 BTC.

Complete Bitcoin Halving History

Four halvings have happened. Each reduced the flow of new coins, and each was followed by a rise in price, though how much the halving itself caused that rise is debated (see below).

Halving Date Block Height Reward Before Reward After BTC Price (Approx.) Price ~1 Year Later
1stNov 28, 2012210,00050 BTC25 BTC$12~$1,000
2ndJul 9, 2016420,00025 BTC12.5 BTC$650~$2,500
3rdMay 11, 2020630,00012.5 BTC6.25 BTC$8,700~$56,000
4thApr 20, 2024840,0006.25 BTC3.125 BTC~$63,800~$85,000

In percentage terms, the price a year after each halving was roughly 8,000%, 280%, 540% and 33% higher. That is not a smooth decline (the third rise was bigger than the second), but the latest was far smaller than any before it. A common reading is that a young, small market can multiply quickly, while a large one moves more slowly. With only four examples, though, any pattern is a weak guide.

The price also didn’t jump on the day of the halving. The halving’s block height is fixed years ahead and its rough date is easy to estimate, so traders can factor it in early; in market language, it may be “priced in.” In past cycles, the largest moves came 6 to 18 months later. One explanation is that miners have fewer new coins to sell, so the selling that weighs on the price (sell pressure) slowly shrinks. Another is that the moves were driven mostly by other forces that happened to coincide. Both views have serious supporters.

Why Bitcoin Has a Fixed Supply of 21 Million

The halving is what makes the cap work. Add up the rewards of every halving era and you get a geometric series:

210,000 × (50 + 25 + 12.5 + 6.25 + 3.125 + ...)

The series 50 + 25 + 12.5 + ... adds up to 100. Multiply by 210,000 blocks per era and you get 21,000,000 BTC.

In practice the total will be slightly less. Bitcoin counts in satoshis (1 BTC = 100,000,000 satoshis) and the software drops any fraction of a satoshi when it halves the subsidy. After the 33rd halving, at block 6,930,000 around the year 2140, the subsidy rounds down to zero and no new bitcoin will be created.

The subsidy counted in satoshis. In January 2009, at block 0, it was 5,000,000,000 satoshis, which is 50 BTC. After 4 halvings, since April 2024, it is 312,500,000 satoshis, 3.125 BTC; after the next halving, around 2028, 156,250,000 satoshis, 1.5625 BTC. After 27 more halvings, at the 32nd, it is 1 satoshi. The 33rd halving would make it half a satoshi, but Bitcoin pays whole satoshis only, so the fraction is dropped and the subsidy becomes 0 from block 6,930,000, around 2140. Because fractions are dropped along the way, the total ends just under 21 million.
The subsidy reaches zero because Bitcoin only counts whole satoshis. Halved 32 times, the original 50 BTC subsidy is down to 1 satoshi. Halved once more it would be half a satoshi, which is dropped, so new coins stop around 2140.

The total passed 20 million around block 940,000, mined on 9 March 2026. In November 2026 about 20.1 million of the 21 million (about 95.8%) have been mined. Some of those coins are lost for good, because their keys are gone. Estimates of lost coins range from 3 to 4 million; nobody knows exactly. See how many bitcoins are there for more.

This schedule is public and predictable. No central bank meeting or political decision can change it, because it is enforced by tens of thousands of independent nodes, each checking every block itself.

How Bitcoin Halvings Affect Mining Economics

For miners, the halving is the most disruptive event in their business cycle. At a block height set years in advance, the subsidy that makes up almost all of their income drops by half while their electricity bill stays the same.

Immediate Revenue Drop

A miner who was barely covering costs before the halving is suddenly losing money. Their machines earn about half as much bitcoin per day. In dollars, hashprice (what a unit of hashing power earns per day, explained in the profitability lesson) also falls by about half on the day, unless bitcoin’s price rises enough to make up for it.

Miner Capitulation

Miners with older, less efficient ASIC machines or expensive electricity are forced to switch off and often sell their machines. This is called miner capitulation.

Difficulty Adjustment Provides Relief

When miners switch off, total hashrate falls, blocks slow down, and at the next difficulty adjustment (every 2,016 blocks) mining becomes easier. The miners who remain each find a larger share of blocks, which softens the blow.

Two charts on the same timeline. Top: the bitcoin a miner who keeps going earns per day. At the halving it is cut in half, while the electricity bill stays the same. Bottom: the network's hashrate. After the halving, older machines and those with costly power switch off, so hashrate falls and blocks slow down. The miner's earnings stay at half until the next difficulty adjustment, every 2,016 blocks, makes mining easier; then the miners left each find more blocks a day and their earnings rise again, which softens the blow.
A halving cuts the bitcoin every miner earns in half overnight; the difficulty adjustment softens the blow. Unless the price rises enough, miners with old machines or expensive power switch off. Once mining gets easier, those who stay each find more blocks.

Hardware Innovation

Halvings reward efficiency. Each one pushes demand toward machines with better joules-per-terahash (J/TH) ratings and pushes the least efficient ones out. The profitability lesson shows the calculation miners use.

Transaction Fees Become More Important

As the subsidy shrinks, transaction fees have to make up a growing share of what miners earn, and in the very long run all of it. That transition has barely started. In the last 1,008 blocks before 6 November 2026 (about a week), fees were about 0.63% of miner revenue; everything else came from the subsidy. Fees spike when many people want block space at once, then fall back.

This raises one of Bitcoin’s most important open questions, called the security budget: will fees alone pay miners enough to keep attacks expensive once the subsidy is tiny? Nobody knows yet. The lesson on transaction fees and Bitcoin’s security budget takes the question further.

Bitcoin Halvings and Price: What the Data Shows

The link between the halving and price is the most argued-over topic in Bitcoin analysis. Every halving so far has been followed by a rise. Whether the halving caused it is another matter.

The Supply Shock Theory

The simplest explanation is supply and demand. Before the 2024 halving the network issued about 900 new BTC a day (144 blocks × 6.25 BTC); afterward, about 450 (144 × 3.125). Miners sell some of their new coins to pay their bills, though how much each day varies and is not the same as what is issued. If demand stays the same or grows, a smaller flow of new coins means upward pressure on the price.

Supporters say the effect builds over months, as fewer new coins reach exchanges and the over-the-counter desks (OTC desks) that arrange large trades away from public order books.

The Stock-to-Flow Model

The stock-to-flow (S2F) model, made popular by the pseudonymous analyst PlanB, tries to put a number on the halving’s effect. It divides the existing stock of bitcoin by the yearly new supply (the flow). Each halving doubles that ratio, and the model linked a higher ratio to a much higher price.

It seemed to fit Bitcoin’s early cycles well, but it has clear limits:

  • It assumes demand keeps growing indefinitely
  • It ignores interest rates, regulation and other conditions
  • Its record is mixed: the original 2019 model’s figure of about $55,000 after the 2020 halving was passed in 2021, but its 2020 follow-up (S2FX) pointed to about $288,000, and bitcoin’s highest price so far is about $126,000 (October 2025)
  • It was fitted to monthly data spanning only two halvings, and just two more have tested it since

Use S2F as a way to think about scarcity, not as a price forecast.

Diminishing Impact

Each halving removes fewer new coins in absolute terms. The 2012 halving cut yearly new supply from about 2.6 million BTC to about 1.3 million. The 2024 halving cut it from about 328,500 BTC to about 164,250. Meanwhile the market is far larger, with ETFs, institutions and derivatives absorbing supply in ways the small market of 2012 could not.

Four bars on the same scale, one per halving, showing new bitcoin created in a year before and after. 2012: from about 2.6 million to about 1.3 million a year, 1.3 million fewer. 2016: from about 1.3 million to about 657,000, 657,000 fewer. 2020: from about 657,000 to about 328,500, 328,500 fewer. 2024: from about 328,500 to about 164,250, 164,250 fewer, a sliver next to the 2012 cut. Each halving removes fewer new coins, and the market is far larger than in 2012.
Each halving takes away fewer new coins than the one before. The 2012 halving cut about 1.3 million new bitcoin a year; the 2024 halving, about 164,250. Meanwhile the market absorbing them has grown far larger.

For a comparison of Bitcoin’s store-of-value properties with other assets, see Bitcoin vs. gold, stocks and fiat.

When Is the Next Bitcoin Halving?

The fifth halving will happen at block 1,050,000, expected around early-to-mid 2028 (around April if blocks keep their pace). The subsidy will drop from 3.125 BTC to 1.5625 BTC. You can watch the countdown on mempool.space.

Halving Estimated Date Block Height Reward After % of Total BTC Mined
5th~20281,050,0001.5625 BTC~96.9%
6th~20321,260,0000.78125 BTC~98.4%
7th~20361,470,0000.390625 BTC~99.2%
33rd~21406,930,0000 BTCAll (just under 21 million)

By 2032 over 98% of all bitcoin will have been mined; the last 2% trickles out over the following century. For what happens after that, read what happens when all bitcoins are mined.

Miners preparing for 2028 are already looking for more efficient hardware and cheaper power. Those in mining pools will see their payout per share fall by about half on the day: the subsidy part halves, while fees don’t.

What the Halving Means for Bitcoin Investors

The halving is a scheduled cut in new supply that has so far been followed by multi-year rises. But “so far” rests on four examples.

A reasonable way to think about it:

  • Don’t try to time the halving. Its date is public, so much of its expected effect may already be in the price.
  • Be wary of “four-year cycle” rules. Many investors plan around the cycle, but the pattern rests on very few cycles, and the latest post-halving rise was much smaller than earlier ones.
  • Know what is certain and what isn’t. The supply schedule is certain. Demand, and therefore price, is not.

Key Takeaways

  • The halving cuts the block subsidy in half every 210,000 blocks (about four years); it is enforced by every node, not by a vote.
  • Four halvings have happened (2012, 2016, 2020, 2024). Each time the price was higher a year later, with the largest moves coming 6 to 18 months after the halving; the latest rise was much smaller than earlier ones.
  • The halving enforces the cap of 21 million bitcoin; about 20.1 million (95.8%) had been mined by November 2026.
  • For miners, each halving cuts income roughly in half overnight, pushing out inefficient machines until difficulty adjusts.
  • Fees were under 1% of miner revenue in late 2026. Whether fees can eventually pay for Bitcoin’s security is the big open question.
  • The next halving, at block 1,050,000 in about 2028, will cut the subsidy from 3.125 to 1.5625 BTC.

Frequently Asked Questions About Bitcoin Halving

Does the bitcoin halving guarantee a price increase?

No. All four halvings were followed by rises, but four examples prove little. The halving reduces new supply; price also depends on demand, interest rates, regulation and sentiment. Each halving also matters less, because most bitcoin has already been issued.

Can the 21 million bitcoin limit ever be changed?

Not by a vote. Every node enforces the supply rule on its own (see who enforces Bitcoin’s rules). To raise the cap, someone would have to release software with new rules, and every user, exchange and business would have to choose to run it. Those who didn’t would reject the extra coins and keep the original Bitcoin, so the change would create a new, separate coin (a hard fork) rather than change Bitcoin for everyone. Because a fixed supply is one of the main reasons people hold bitcoin, such a fork is widely expected to get little support.

What happens to Bitcoin’s security after all 21 million are mined?

Miners will be paid only by transaction fees. The shift happens gradually, one halving at a time. Whether fees alone can pay for enough security is an open, actively debated question, especially since fees were under 1% of miner revenue in late 2026; the security budget lesson covers it in depth. Layer 2 systems like the Lightning Network move many payments off the main chain, but they still need on-chain transactions to open and close channels, so their effect on fees can go either way.

Why did Satoshi choose 21 million as the cap?

Satoshi never fully explained it. The number follows directly from the 50 BTC starting reward and the 210,000-block halving interval. One common theory is that 21 million coins, each divisible into 100 million satoshis, gives 2.1 quadrillion units, enough for a worldwide currency.

Do I need to do anything during a halving?

If you hold bitcoin: nothing. Your coins aren’t affected. If you mine: check well before the date whether your operation still covers its costs at half the subsidy, and plan for more efficient machines or cheaper electricity if it doesn’t.

Check Your Understanding

1. A miner’s software is changed to pay itself 6.25 BTC plus fees per block today. What happens to that block?

Answer

Every full node rejects it, because the rules say the subsidy in this era is 3.125 BTC (plus fees). The block is invalid however much work went into it, and the miner loses the reward. That is how the schedule is enforced: by each node checking, not by a vote.

2. Before 2024 the network issued about 900 BTC a day; after, about 450. Does that mean miners sell 450 fewer BTC a day?

Answer

No. Those are the numbers of new coins created. Miners sell some of them, and sometimes coins they held from earlier, to pay their bills. How much they actually sell varies. The halving halves the new supply, not necessarily the selling.

3. Why does the subsidy reach zero at the 33rd halving and not “never”?

Answer

Bitcoin counts in whole satoshis. Halving 50 BTC (5 billion satoshis) 32 times leaves 1 satoshi; halving it once more gives half a satoshi, which the software rounds down to 0. That happens at block 6,930,000, around the year 2140.