The Bitcoin Halving Explained: Impact on Mining

In short. Roughly every four years, the Bitcoin block reward is cut in half. Four halvings have occurred since 2009, from 50 down to 3.125 BTC. The fifth is expected in spring 2028, at 1.5625 BTC. For a miner, each halving forces inefficient hardware offline and durably compresses hashprice. A 2026 ASIC purchase amortized over twenty-four months crosses that date, which makes modelling it mandatory.

The mechanism

Every 210,000 blocks, roughly four years at the ten-minute target, the reward a miner receives for a validated block is cut in half. The rule is written into the protocol and cannot change without a hardfork the entire network accepts.

Why it exists

Bitcoin was designed with a money supply capped at 21 million units. The halving progressively slows issuance to respect that cap without ever reaching it abruptly.

BTC annual inflation follows mechanically:

Period	Block reward	Annual inflation
2020 to 2024	6.25 BTC	about 1.8 %
2024 to 2028	3.125 BTC	about 0.8 %
2028 to 2032	1.5625 BTC	about 0.4 %

For comparison, monetary inflation in major currencies generally runs between 2 and 8 % per year.

The four past halvings

Halving	Date	Reward	Hashrate before	BTC price at the date
1	November 2012	50 to 25 BTC	about 25 TH/s	about $12
2	July 2016	25 to 12.5 BTC	about 1.5 EH/s	about $650
3	May 2020	12.5 to 6.25 BTC	about 120 EH/s	about $9,000
4	April 2024	6.25 to 3.125 BTC	about 620 EH/s	about $64,000

BTC price rose meaningfully in the twelve to eighteen months following each of those events. Four occurrences do not make a statistical model, and nothing commits the fifth to behaving like its predecessors.

Three concrete effects on operations

BTC revenue is cut in half

At the exact moment of the halving, a miner's BTC revenue halves at constant hashrate. A machine producing 0.001 BTC per month the day before will produce 0.0005 the day after.

The only catch-up runs through a price increase that offsets the reduced issuance. That is what happened historically, without any mechanism guaranteeing it.

Inefficient machines leave the network

Miners running high J/TH hardware, such as S19 units around 27.5 J/TH, become immediately cash-negative after the halving and unplug.

The effect cascades: global hashrate drops, difficulty adjusts downward at the next retarget, and efficient miners see profitability partially recover. That is the network's self-regulation mechanism, and it has repeated at every halving since 2012.

Hashprice stays durably compressed

Hashprice, meaning revenue in dollars per terahash per day, mechanically halves at the event, then partially recovers through the difficulty adjustment. It nevertheless stays structurally below the previous year for six to twelve months.

That compression, more than the halving itself, is what eliminates marginal operators and concentrates profitability on the most efficient fleets.

The next halving

At ten minutes per block and 210,000 blocks between halvings, block 1,050,000 will trigger the fifth, expected between late April and mid-May 2028. The reward will then fall from 3.125 to 1.5625 BTC per block.

If you buy an ASIC in 2026 aiming at a twenty-four month payback, your model crosses that date. At constant hashprice, your BTC revenue halves that day, absent a compensating price move.

Three practical consequences:

• Target efficiency below 12 J/TH, which limits the choice to hydro models such as the S23 Hyd 3U at 9.5 J/TH or the S21 XP Hyd at 12 J/TH.
• Secure an electricity cost below $0.08/kWh, a level almost no residential tariff reaches, hence the move to professional hosting.
• Compute the post-halving break-even before buying, with the reward already halved.

The calculation to run before buying

Take your break-even kWh and halve the hashprice. An S21 Pro, at 15 J/TH and 84.2 kWh per day, earns $11.70 per day at a hashprice of $50/PH/day, giving a break-even of $0.139/kWh. After the halving, at half the hashprice and an unchanged BTC price, that revenue falls to $5.85 and the break-even to $0.069/kWh.

In other words, a machine perfectly viable today at a host charging $0.075/kWh becomes loss-making overnight, with no failure having occurred. That is precisely the calculation most purchase plans omit.

Run it with three explicit assumptions: the future 1.5625 BTC reward, a global hashrate 30 % above today's, and an unchanged BTC price. If the project holds under that scenario, it will probably hold under the others.

A miner's three levers

Optimize J/TH. Refresh hardware every three to four years. S19 units from 2022, around 27.5 J/TH, already sit at break-even in 2026. S21 units at 15 J/TH will hold through the 2028 halving but not beyond at most rates. S23 units at 9.5 J/TH clear the threshold.

Cut the price per kWh. This is by far the most powerful lever. On an S21 Pro drawing 30,748 kWh per year, moving from $0.16/kWh, a common US residential rate, to $0.07/kWh in hosting saves $2,767 per year. That is not a margin improvement: it is what moves the machine from losing $3.46 a day to earning $5.80.

Model before buying, not after. See the ASIC miner ROI calculator for the full method.

Further reading

• Is bitcoin mining profitable in 2026?
• Bitcoin mining electricity cost
• Hosted bitcoin mining

Any projection about the 2028 halving depends on assumptions that will change. Mining involves a risk of capital loss.