Every 210,000 blocks, Bitcoin's code cuts the number of new coins paid to miners in half. Nobody votes on it and nobody announces a date; the cut takes effect automatically when the block count reaches the next multiple of 210,000. This guide sets out the schedule from the source code, every past halving with its block height and date from public block data, what the April 2024 cut did to miners, when the next one should arrive, and what the price record can and cannot tell you. For how the network itself works, see what is bitcoin; the crypto hub covers the wider market.
What is a bitcoin halving?
A halving is the block at which the subsidy, the newly created bitcoin each block may pay its miner, drops by half.
The rule takes a few lines of Bitcoin Core's GetBlockSubsidy function. It divides the block height by 210,000, drops the remainder, and shifts the starting subsidy of 50 BTC right by that many binary places, which halves it each time. A comment beside the code says the cut "will occur approximately every 4 years."
Subsidies are counted in whole satoshis, 100 million to a bitcoin, so repeated halving eventually rounds them away: the subsidy shrinks to 1 satoshi at block 6,720,000 and to zero at block 6,930,000, around 2140. That rounding is why the total stops at just under 21 million BTC (20,999,999.9769 BTC) rather than at exactly 21 million. Nothing else changes at a halving. The 10-minute target, the difficulty rules and the fee market carry on as before, and only the miners' income in bitcoin drops overnight.
When were the past bitcoin halvings?
Bitcoin has halved four times, on November 28, 2012, July 9, 2016, May 11, 2020 and April 20, 2024, with all dates in UTC.
| Event | Block height | Date and time (UTC) | Subsidy per block | Issued before this block | Fees in the block itself |
|---|---|---|---|---|---|
| Launch | 0 | January 3, 2009, 18:15 | 50 BTC | 0 BTC | none |
| First halving | 210,000 | November 28, 2012, 15:24 | 50 to 25 BTC | 10,500,000 BTC (50%) | 13.56 BTC |
| Second halving | 420,000 | July 9, 2016, 16:46 | 25 to 12.5 BTC | 15,750,000 BTC (75%) | 0.58 BTC |
| Third halving | 630,000 | May 11, 2020, 19:23 | 12.5 to 6.25 BTC | 18,375,000 BTC (87.5%) | 0.91 BTC |
| Fourth halving | 840,000 | April 20, 2024, 00:09 | 6.25 to 3.125 BTC | 19,687,500 BTC (93.75%) | 37.63 BTC |
| Fifth halving | 1,050,000 | Expected around April 2028 | 3.125 to 1.5625 BTC | 20,343,750 BTC (96.875%) | not yet known |
The four-year rhythm is approximate. Halvings are set by block height, and blocks come faster than every 10 minutes whenever hash rate is growing, because difficulty only catches up at each two-week reset. The gaps between halvings were 1,319, 1,402 and 1,440 days, against the 1,458 days that 210,000 blocks would take at exactly 10 minutes, which means blocks averaged 9.04, 9.61 and 9.87 minutes in those eras. Percentages in the table are shares of the final supply.
What changed at the April 2024 halving?
Block 840,000, mined at 00:09 UTC on April 20, 2024, the evening of April 19 in US time zones, cut the subsidy from 6.25 to 3.125 BTC.
New supply fell from 900 to 450 bitcoins a day. The same block launched Runes, a protocol for creating tokens on Bitcoin, and users racing to create the first ones paid 37.63 BTC in fees in that single block, about 12 times its new coins, at a median rate of 200 sat/vB. NYDIG's research team noted that less than a week later, fee rates had fallen below their pre-halving levels.
Miners did not leave in a rush. The first difficulty reset after the halving, on April 24, 2024, raised difficulty 1.99%, and the next, on May 9, cut it 5.63% as less efficient machines were switched off. Growth then resumed: the network briefly touched 1 zettahash per second, or 1,000 exahashes, several times in 2025 and first held that level on a seven-day average on September 2, 2025.
The order of events in price was new as well. Coinbase's daily close passed its 2021 record on March 4, 2024, 47 days before the halving, whereas in the previous cycle the 2017 record was not beaten until November 30, 2020, more than six months after that cycle's halving. The close on halving day was $64,905.95.
When is the next bitcoin halving?
Block 1,050,000 triggers the fifth halving, and at recent block speeds it should arrive around the second week of April 2028.
Worked example: Estimating the date of the fifth halving
Start from the newest block when this guide was checked, then divide the remaining blocks by an assumed pace.
- Block 974,194 was mined at 00:39 UTC on November 2, 2026.
- Blocks still to go: 1,050,000 − 974,194 = 75,806.
- At exactly 10 minutes each: 758,060 minutes, or 526.4 days, which lands on April 11, 2028.
- At the 9.94-minute average since block 840,000: about 523 days, or April 8, 2028.
- If blocks sped up to 9.5 minutes, the date would move forward to about March 16, 2028; if they slowed to 10.5 minutes, it would slip to about May 7, 2028.
The schedule therefore puts the next cut at block 1,050,000, expected around April 2028, when the subsidy falls to 1.5625 BTC. New supply will then drop to about 225 bitcoins a day, or 82,125 a year, roughly 0.40% of the coins in existence at that point. Countdown clocks on block explorers use the same arithmetic and update with each new block, so expect any published date to drift by days or weeks.
How does the halving affect miners?
A halving cuts miners' income in bitcoin by close to half overnight, while their electricity and hardware bills stay the same.
Miners earn the subsidy plus fees, and fees have been a small and uneven share. Averaging mempool.space's block data by calendar year gives the approximate figures below; the peaks in 2017 and 2023 to 2024 came when congestion and new token protocols filled blocks, and you can read about those in the explainers on Bitcoin Ordinals and BRC-20 tokens.
| Year | Average fees per block | Fees as a share of miner revenue |
|---|---|---|
| 2017 | 1.84 BTC | 12.8% |
| 2020 | 0.52 BTC | 5.7% |
| 2021 | 0.42 BTC | 6.3% |
| 2022 | 0.10 BTC | 1.6% |
| 2023 | 0.44 BTC | 6.5% |
| 2024 | 0.29 BTC | 6.6% |
| 2025 | 0.03 BTC | 1.0% |
| 2026, through November 2 | 0.02 BTC | 0.6% |
What keeps the system stable is difficulty. If enough machines switch off, blocks slow until the next reset makes mining easier for the survivors. The largest cut on record came on July 3, 2021, when difficulty fell 27.94% after China's crackdown pushed miners offline; CoinDesk reported that hash rate had averaged 87.7 exahashes a second in the period before the reset, down from about 180 in mid-May. The figure miners watch is hashprice, the revenue per unit of computing power: over the 144 blocks to November 2, 2026, the network paid 452.11 BTC while running at about 1,046 exahashes a second, or about 0.000432 BTC per petahash a second per day.
Worked example: What the 2028 halving would do to two miners at today's numbers
Hold today's hash rate and fees constant, assume a bitcoin price of $80,000 and power at $0.05 per kilowatt-hour, and measure everything per petahash a second (1,000 terahashes) of mining capacity.
- Revenue now: 0.0004322 BTC × $80,000 = $34.58 a day.
- Miner A runs machines using 16 joules per terahash, in line with the 16.0 J/TH fleet average one Nasdaq-listed miner reported in April 2026: 16 kilowatts × 24 hours = 384 kWh, costing $19.20 a day, for a margin of $15.38.
- Miner B runs older machines at an assumed 30 joules per terahash: 720 kWh, costing $36.00, for a margin of −$1.42.
- After the halving, daily network revenue falls to about 225 BTC of subsidy plus 2.11 BTC of fees, or 0.0002171 BTC per petahash, which is $17.37 a day. Miner A's margin becomes −$1.83 and Miner B's −$18.63.
At those numbers something has to give: the price rises, fees rise, or enough machines switch off that difficulty falls and each remaining petahash earns more. Which of these happens cannot be known in advance.
Do halvings push the bitcoin price up?
The record shows big rallies after each of the last three halvings, but three cases cannot separate the halving's effect from everything else happening at the time.
The table uses Coinbase's daily closing prices from FRED, the St. Louis Fed's database. The series starts in December 2014, so it does not cover the 2012 halving.
| Halving | Close on halving day | Close 12 months later | Highest close within two years | Days from halving to that high | Fall from that high to the next low |
|---|---|---|---|---|---|
| July 9, 2016 | $657.61 | $2,508.99 (+282%) | $19,650.01 on December 16, 2017 | 525 | −83.8% by December 15, 2018 |
| May 11, 2020 | $8,557.37 | $56,719.00 (+563%) | $67,510.06 on November 8, 2021 | 546 | −76.7% by November 21, 2022 |
| April 20, 2024 | $64,905.95 | $87,522.07 (+35%) | $124,720.09 on October 5, 2025 | 533 | −53.0% by June 30, 2026 |
What the data shows:
- Each of the three halvings in the series was followed, 17 to 18 months later, by the cycle's highest close, and then by a fall of more than half.
- The rallies shrank. The peak was 29.9 times the halving-day price in the 2016 cycle, 7.9 times in the 2020 cycle and 1.9 times in the 2024 cycle.
- The supply shock shrank too. The 2016 cut reduced yearly issuance by 4.17% of existing supply, the 2020 cut by 1.79% and the 2024 cut by 0.83%; the 2028 cut will remove 0.40%.
What it does not show:
- Cause. Three data points cannot rule out other drivers, and each cycle had big ones: the 2017 boom in new token sales, near-zero interest rates and pandemic stimulus in 2020 and 2021, and the launch of US spot bitcoin ETFs in January 2024.
- Surprise. Halvings are scheduled years ahead, and the 12 months before each one rose as well, by 144% before 2016, 19% before 2020 and 130% before 2024.
- A timing rule. This guide makes no forecast, and the pattern is not a trading signal. The crypto market cycles guide and the bitcoin drawdowns study look at cycles in more depth, and the bitcoin price page shows the latest market data.
How much bitcoin will exist after future halvings?
By the schedule, 20,343,750 BTC will exist when the 2028 halving begins, and 99% of the final supply will have been issued by about 2035.
Worked example: Supply at the fifth halving and what is left after it
- Coins issued by the first four eras, from the table above: 19,687,500 BTC.
- Coins the current era adds, blocks 840,000 to 1,049,999: 210,000 × 3.125 = 656,250 BTC.
- Supply when block 1,050,000 arrives: 19,687,500 + 656,250 = 20,343,750 BTC, or 96.875% of the maximum.
- Coins left for every later era combined: 20,999,999.9769 − 20,343,750 ≈ 656,250 BTC, almost exactly what the current era alone releases.
- Those remaining coins trickle out over about 112 years, from 2028 to around 2140.
That symmetry follows from halving: each era issues as much as all later eras put together. The next few eras look like this, with years approximate.
| Era begins at block | Around | Subsidy per block | New bitcoin per day | Supply at the end of the era | Share of the maximum |
|---|---|---|---|---|---|
| 1,050,000 | 2028 | 1.5625 BTC | 225 | 20,671,875 BTC | 98.44% |
| 1,260,000 | 2032 | 0.78125 BTC | 112.5 | 20,835,937.5 BTC | 99.22% |
| 1,470,000 | 2036 | 0.390625 BTC | 56.25 | 20,917,968.75 BTC | 99.61% |
| 1,680,000 | 2040 | 0.1953125 BTC | 28.125 | 20,958,984.375 BTC | 99.80% |
| 6,930,000 | 2140 | 0 | 0 | 20,999,999.9769 BTC | 100% |
These are issuance figures. Coins that are lost, or locked in wallets whose keys are gone, still count, so the supply anyone can actually spend is lower.
What are the risks of reading too much into the halving?
The biggest risk is treating a supply rule as a price signal: bitcoin fell by more than half after each of the last three post-halving peaks.
Price risk. A buyer at the 2017, 2021 or 2025 highs sat through losses of 83.8%, 76.7% and 53.0% on daily closes. Crypto can lose most or all of its value, and a halving changes none of that.
Hype and scams. Halving countdowns are a favorite hook for promotions, including offers to multiply any coins you send to an address, which are always scams; crypto scams lists the common patterns.
A shrinking security budget. The subsidy that pays miners to protect the chain halves again in 2028. Unless fees or the price make up the difference, fewer machines will be paid to do that work, and with fees at roughly half a percent of miner revenue in the month to November 2, 2026, the question is open.
Miner stress. When margins turn negative, miners with debt or expensive power may have to sell coins or shut down, and blocks can slow until the next difficulty reset. If only the largest, most efficient operators survive each cut, block production also becomes more concentrated.
For the practical side of owning bitcoin, see how to buy bitcoin and how much crypto to hold; the comparison of bitcoin vs ethereum explains how ether's issuance works without halvings.