The Bitcoin Halving, Explained: Supply, Miners, and the Price Cycle

Every 210,000 blocks, Bitcoin cuts the reward it pays miners in half. That single rule is the halving, and it is the reason new bitcoin arrives on a schedule fixed in 2009 rather than whenever someone decides they want more of it. The rule lives in consensus code that every full node enforces, so no company, foundation, or lead developer can quietly override it. What follows is how the halving actually works, when the next one lands, what it does to the people running the machines, and whether the famous four-year price cycle is a real pattern or just four data points wearing a costume.

How it works

The block subsidy is the newly minted bitcoin paid to whoever mines each block. It started at 50 BTC when the genesis block was mined on January 3, 2009. Every 210,000 blocks, the protocol cuts that subsidy in half. At the network's 10-minute average block target, 210,000 blocks take roughly four years to produce, which is where the loose "every four years" shorthand comes from.

Nobody votes on this. It is arithmetic baked into the consensus rules, and every full node checks it independently. A block that tries to pay its miner more than the schedule allows is not up for debate, it is simply invalid, and the rest of the network throws it out. That is the entire mechanism: scarcity enforced by software running on thousands of independent machines that all agree on the same math.

That design is what makes Bitcoin's issuance different in kind from most other money. Supply here does not respond to price, to demand, or to anyone's decision. It follows the block count, and nothing else.

One common confusion is worth clearing up. The halving does not change how fast blocks arrive. Difficulty adjusts to keep blocks landing about every 10 minutes no matter how much mining power joins or leaves. A halving changes how many new coins each block creates, not the pace at which blocks come.

Four cuts so far

Bitcoin has halved four times. Block 210,000, on November 28, 2012, took the subsidy from 50 to 25 BTC. Block 420,000, on July 9, 2016, dropped it to 12.5. Block 630,000, on May 11, 2020, brought it to 6.25. Block 840,000, on April 20, 2024, cut it to 3.125 BTC, where it has stood since.

The pattern keeps running. The subsidy will halve 32 times in total, each cut smaller in absolute terms than the one before, until sometime in the early 2140s the reward rounds down to zero satoshis. After that, miners earn transaction fees and nothing else.

When the next one lands

The next halving comes at block 1,050,000, when the subsidy drops from 3.125 to 1.5625 BTC. Block-target math puts that around 2028. The date is an estimate, though, not a fixed entry on a calendar.

Difficulty adjustment is the reason. Every 2,016 blocks, roughly two weeks, the protocol retunes mining difficulty to hold the 10-minute average as hashrate rises and falls. Real block times still drift slightly above or below target, so the exact moment block 1,050,000 gets mined can only be projected from the current pace. That is why every halving countdown, including the live one on cryptonews.gg/tools/bitcoin-halving, is an estimate that tightens as the block draws near rather than a firm date.

What it does to miners

A miner's revenue is the subsidy plus whatever fees the transactions in its block are paying. A halving cuts the subsidy side in half overnight while the power bill and the hardware financing stay exactly where they were. The squeeze lands on the least efficient machines first.

What usually follows is a shakeout. Some hashrate goes dark as unprofitable rigs power down. The next difficulty adjustment then ratchets down to match the smaller network, and the miners still running see their margins partly recover. None of this catches the professionals off guard: public mining companies prepare for halvings well in advance, swapping in more efficient ASICs and locking in cheaper power contracts, then leaning on cash reserves to survive the lean stretch right after the cut.

The recovery is neither guaranteed nor instant. Difficulty only retunes every 2,016 blocks, so there can be a brutal window between the subsidy cut and the adjustment that eases it. The miners who walked in with thin margins and expensive power are the ones who do not always walk out.

Fees are the long game. In calm markets they are usually a low single-digit share of miner revenue, but they can spike violently when demand surges. The April 2024 halving handed everyone a strange preview. Block 840,000, the halving block itself, carried around 37.6 BTC in fees, many times its brand-new 3.125 BTC subsidy, because the Runes protocol launched at exactly that block and set off a rush of transactions. The spike faded within days. For a moment it put on display the future Bitcoin's designers are counting on, one where fees rather than fresh coins pay for the network's security.

The 21 million cap

The halving schedule is the machinery behind Bitcoin's 21 million coin cap. Sum every subsidy across all 32 halvings and total issuance converges just under that round number, at roughly 20,999,999.97 BTC. Most of it is already out in the world. By mid-2024, more than 19.7 million BTC had been mined, over 93 percent of the coins that will ever exist.

There is a counterintuitive point buried in those numbers. Because so much of the supply is already mined, each remaining halving removes a smaller and smaller absolute amount of new bitcoin from the schedule. In raw coin terms, the dramatic cuts are behind us. What is left is a long, slow taper toward zero.

The flow of new supply is already thin. After the April 2024 halving, Bitcoin's annual issuance fell to about 0.85 percent of existing supply, below the 1 to 2 percent range usually cited for how fast the world's gold stock grows each year. Every future halving nudges that figure closer to nothing.

When the subsidy finally rounds to zero in the 2140s, miners get paid entirely in fees. Whether fee demand will be high enough and steady enough to keep the network secure by then is a real open question, argued among researchers under the heading of Bitcoin's long-term security budget. It is unsettled, and it does not have to be settled for another century.

Does it move the price

This is the part most people actually came for, and it is where the evidence is thinnest. The story runs like this. Bitcoin traded near 12 dollars at the 2012 halving and reached roughly 1,150 dollars about a year later. It sat near 650 dollars in July 2016 and peaked around 19,700 dollars some 17 months on. It was near 8,700 dollars in May 2020 and topped out near 69,000 dollars about 18 months after that. At the April 2024 halving it changed hands near 64,000 dollars. A cycle top 12 to 18 months past each halving: that is the four-year-cycle narrative in a single breath.

The objections are not nitpicks. The whole sample is four events. Each of those cycles also overlapped with something enormous that had nothing to do with block rewards, from the monetary stimulus of 2020 and 2021 to the launch of US spot Bitcoin ETFs in 2024. Returns have shrunk with every cycle.

The priced-in objection deserves its own beat. The halving is not a surprise. Its block height has been public since the start, and anyone who wanted to position for it has had years to do so. A market that knows exactly when new supply will drop should, in theory, have moved on that information well before the block is mined. If the halving still reliably jolted price on the day, that would be a liquid market leaving money on the table, which is not how liquid markets tend to behave.

The stock-to-flow model ran headlong into this. It tried to predict price straight from the ratio of existing supply to new issuance, drew an enormous following after the pseudonymous analyst PlanB popularized it in 2019, then missed badly through 2021 and 2022. Quantitative researchers now largely dismiss it.

So here is the honest version. A halving mechanically cuts the flow of new supply. That much is certain. Past cycles have loosely tracked the halvings. But nobody has demonstrated the link is causal, and nobody has shown it will repeat. The next data point arrives at block 1,050,000, around 2028.