In mid-September 2026, the Bitcoin network passed block 967,751. By then, 95.65% of the 21 million bitcoins were already in circulation: 19,687,500 BTC had been created before April 2024, another 127,751 blocks at 3.125 BTC added 399,222 more, and the remaining 4.35% will trickle out until around 2140 (¹). Seventeen years to issue almost everything. More than a century for the rest. That imbalance has a name: the halving. You have probably seen the word in a headline, often next to “boom” or “scarcity”. But what actually happens on that day? Who decides it? And why do only a handful of cryptocurrencies use it? Here is everything you need to know, with the maths to back it up.
Key takeaways
📋 Index
- A halving cuts the reward miners receive for each new block in half.
- On Bitcoin, it happens every 210,000 blocks, a little under four years.
- The last one took place on 20 April 2024 (3.125 BTC per block); the next is expected around April 2028.
- Litecoin, Bitcoin Cash and Zcash use the same principle, each at its own pace.
- It slows the creation of new coins, but never guarantees a price rise.
Crypto halving, a cut in half written into the protocol
The word comes straight from “half”. The idea fits on one line: at a fixed interval, the protocol takes the scissors to the paycheck, and miners receive half as many new coins for the same work.
The block reward, the salary the halving trims
On a blockchain secured by proof of work, machines compete to validate the next block of transactions. The winner collects two things: the fees paid by users and a subsidy in freshly minted coins. Only the subsidy is halved. The fees stay untouched.
Picture a tap. Each block lets a trickle run. Roughly every four years, an invisible hand turns it halfway shut. The water never stops all at once. It slows down in steps, like a staircase whose steps keep shrinking.
Whose hand is it? Nobody’s. No central bank meets, no committee votes. Every computer on the network applies the same rule at the same block, and any node that deviated would simply be ignored by the others.
But the most surprising part hides in a few lines of code.
Under the hood: a single binary operation
You would expect a sophisticated formula. Wrong. The halving mechanism boils down to an instruction every computer science student knows.
210,000 blocks, not a date
First key point: the trigger is a block height, never a calendar day. Bitcoin’s first halving fell at block 210,000, the second at block 420,000, and so on. That is why no two websites give you exactly the same date. Each one extrapolates from the recent speed of block production.
What Bitcoin’s code actually says
In the reference software, Bitcoin Core, the function that computes the subsidy starts from 50 BTC expressed in satoshis, the smallest unit: 5,000,000,000. It divides the block height by 210,000 to count past halvings. Then it shifts that number to the right in binary, by as many positions as there have been halvings. Shifting one position equals dividing by two, remainder discarded.
A rarely discussed consequence: after 32 halvings, the reward drops to a single satoshi, and the 33rd brings it to zero, at block 6,930,000 (²). These successive roundings also nibble at the total. The real cap therefore never quite reaches 21 million. It peaks at 20,999,999.9769 BTC.
One safeguard completes the picture. If the number of halvings reaches 64, the code returns zero directly, because such a large shift would produce an undefined result in C++. A programmer’s precaution. It will never matter in practice, since zero arrives at the 33rd step.
One practical question remains. When do these steps land?
How often? Every four years… in theory
Bitcoin targets one block every ten minutes. The maths looks simple: 210,000 blocks × 10 minutes, or 1,458 days. Yet no cycle has ever matched that figure.
| Halving | Date | Block | Reward | Days since previous step |
|---|---|---|---|---|
| Launch | 3 January 2009 | 0 | 50 BTC | — |
| 1st | 28 November 2012 | 210,000 | 25 BTC | 1,425 |
| 2nd | 9 July 2016 | 420,000 | 12.5 BTC | 1,319 |
| 3rd | 11 May 2020 | 630,000 | 6.25 BTC | 1,402 |
| 4th | 20 April 2024 | 840,000 | 3.125 BTC | 1,440 |
| 5th (expected) | April 2028 | 1,050,000 | 1.5625 BTC | — |
Why is it always early? Because mining difficulty only readjusts every 2,016 blocks. Between two adjustments, if new machines join the network, blocks arrive faster than planned. The protocol always corrects with a delay.
The gap from the theoretical 1,458 days keeps shrinking from one cycle to the next: 139 days early between 2012 and 2016, 56 between 2016 and 2020, then only 18 between 2020 and 2024 (³). While computing power was exploding, blocks raced ahead of the target. As its growth becomes steadier, the calendar moves closer to the ideal pace. Current estimates, which place the next step around 13-17 April 2028, confirm this tightening.
The calendar is one thing. But why design it at all?
Programmed scarcity, an idea born from a banking crisis
Bitcoin’s first block carries almost nothing. Just a 50 BTC reward and a sentence slipped in by its creator, Satoshi Nakamoto, taken from that day’s front page of the British newspaper The Times:
« The Times 03/Jan/2009 Chancellor on brink of second bailout for banks. »
A timestamp, and a manifesto. In the middle of a financial crisis, governments were bailing out banks by creating money. Satoshi proposed the opposite: issuance known in advance, impossible to speed up, shrinking until it fades out. The halving is the tool behind that promise.
Inflation that halves every four years
Let’s do the maths. With 144 theoretical blocks per day and 3.125 BTC per block, the network creates 450 BTC daily. Over a year, that adds up to 164,250 BTC, a monetary inflation of about 0.82% relative to existing coins. After 2028? 225 BTC per day, and a rate close to 0.40% (⁴). For comparison, the European Central Bank targets 2% price inflation: the two concepts differ, but the order of magnitude speaks for itself.
Hence the nickname “digital gold”. Gold also gets harder to extract over time. The difference? Nobody can discover a new bitcoin deposit.
And after 2140? Miners will live on transaction fees alone. Nobody can yet prove that bet will pay off.
To estimate the next halving date yourself, ignore the conflicting countdowns. Check the latest block height on any public explorer, subtract it from 1,050,000, then multiply by 10 minutes. At block 967,751, 82,249 blocks remained, roughly 571 days. Redo the calculation a month later: the difference will tell you whether the network is speeding up or slowing down.
Bitcoin, however, does not have a monopoly on the process.
Bitcoin, Litecoin, Zcash: the halving club
Most of the projects involved descend from Bitcoin’s code. They inherited the mechanism, sometimes adjusting the pace.
| Crypto | Interval | Target block time | Current reward | Next halving | Max supply |
|---|---|---|---|---|---|
| Bitcoin (BTC) | 210,000 blocks | 10 min | 3.125 BTC | April 2028 | 21 million |
| Litecoin (LTC) | 840,000 blocks | 2.5 min | 6.25 LTC | summer 2027 | 84 million |
| Bitcoin Cash (BCH) | 210,000 blocks | 10 min | 3.125 BCH | 2028 | 21 million |
| Zcash (ZEC) | 1,680,000 blocks | 75 s | 1.5625 ZEC | late 2028 | 21 million |
Litecoin produces four times as many blocks, but over an interval four times longer. Result: the same four-year rhythm. Its halvings fell in August 2015, August 2019 and August 2023, at block 2,520,000. Today, 576 daily blocks at 6.25 LTC inject 3,600 LTC per day (⁵).
Bitcoin Cash, born from a Bitcoin split in 2017, went through its halvings in April 2020 and early April 2024, a few days ahead of its big brother. Why so early? For a while after its creation, its blocks came in faster. Bitcoin SV, itself a spin-off of Bitcoin Cash, follows the same 210,000-block interval.
Zcash adds a twist. Since its November 2024 upgrade, its 1.5625 ZEC reward no longer goes entirely to miners: 80% to them, 8% to a community grants fund, 12% into a locked reserve. In practice, 1,152 daily blocks create 1,800 ZEC, of which only 1,440 go to mining.
The coins that turned their back on halving
Ethereum? No halving. It moved to proof of stake in September 2022, and its issuance follows other rules. Dogecoin has frozen its reward at 10,000 DOGE per block since 2015: an uncapped supply, by design. Monero chose a “tail emission” of 0.6 XMR per block since 2022, to pay for network security forever. Kaspa, finally, went for a gentle slope: its reward drops a little every month, so that it halves over a full year, with no sudden step.
Three philosophies, three bets. And one question everyone asks.
Price and miners: what the halving really changes
For miners, the effect shows up from the very next block. Their income in new coins drops by half overnight, while their electricity bills do not move a cent. The least efficient rigs become unprofitable and switch off, unless the price or fees make up the difference.
On price, reality is far more fickle than the legend. Bitcoin Cash’s April 2024 halving was followed by a 10% jump within the day. Litecoin’s in August 2023? A drop of nearly 8% in twelve hours (⁶). Same mechanism, opposite reactions. A halving never trades in a vacuum: interest rates, regulation and risk appetite weigh just as much.
Believing the halving cuts the number of bitcoins in circulation in half, or mechanically doubles the price. It only affects the flow of new coins, never the existing stock. At the last halving, 19,687,500 BTC were already in circulation and not a single one vanished. Only daily production fell, from 900 to 450 BTC. The mistake comes from the word itself: “half” suggests a clean cut in total supply, when it is really just a tap being tightened.
So should you watch the next halving as a buy signal? The honest answer fits in two words: not necessarily. The event is known years in advance. Markets have plenty of time to anticipate it, and often to digest it before it happens. What the halving does guarantee, on the other hand, never changes: a supply more predictable than any currency issued by a central bank.
Sources and statistics
(1) Before the April 2024 halving, four eras of 210,000 blocks had issued 210,000 × (50 + 25 + 12.5 + 6.25) = 19,687,500 BTC. The CoinWarz tracker showed block 967,751 in mid-September 2026, meaning 127,751 blocks at 3.125 BTC since block 840,000, or 399,221.875 additional BTC. Total: 20,086,722 BTC, i.e. 95.65% of 21 million. That leaves 913,278 BTC to be issued until around 2140.
(2) According to Bitcoin Core’s subsidy function, 50 BTC equals 5,000,000,000 satoshis. Since 232 = 4,294,967,296 is below that number and 233 is above it, 32 binary shifts leave 1 satoshi and the 33rd gives zero. That 33rd halving occurs at block 33 × 210,000 = 6,930,000.
(3) The gaps are calculated in days between the official halving dates (28/11/2012, 09/07/2016, 11/05/2020, 20/04/2024), then compared with the theoretical duration of 210,000 × 10 minutes = 1,458 days. This gives 1,319, 1,402 and 1,440 days, i.e. 139, 56 and 18 days early. Estimates for the next halving come from CoinGecko and CoinWarz.
(4) Daily issuance: 6 blocks per hour × 24 = 144 blocks, × 3.125 = 450 BTC. Annual: 450 × 365 = 164,250 BTC, divided by the 20,086,722 BTC in existence, giving 0.82%. After 2028, 225 BTC per day for about 20.34 million in circulation, giving 0.40%.
(5) Litecoin’s halving dates and heights (blocks 840,000, 1,680,000 and 2,520,000) are confirmed by BitPay and the F2Pool mining pool. At 2.5 minutes per block, there are 576 blocks per day, × 6.25 = 3,600 LTC. For Zcash, the 80/8/12 split and the 1.5625 ZEC reward come from announcements tied to the NU6 upgrade; 86,400 seconds ÷ 75 = 1,152 blocks, × 1.5625 = 1,800 ZEC.
(6) Bitcoin Cash’s 10% rise after its 4 April 2024 halving was reported by CoinDesk. Litecoin’s drop of nearly 8% within twelve hours of its 2 August 2023 halving was noted by CoinDCX.
