Daily Issuance

The chart stacks the three things that fall out of Bitcoin’s issuance schedule. The top panel is the BTC count per day — a clean step function on a logarithmic axis, with vertical drops at 2012-11-28, 2016-07-09, 2020-05-11, and the April 2024 halving. The log axis is doing deliberate work: it places the early 7,200 BTC/day era on the same visual band as today’s 450, so all four halvings read at equal amplitude. On a linear axis the 2010 era would dwarf everything after 2016 and the step structure would vanish.

Beneath it is the identical flow multiplied by the daily close, and alongside is the BTC flow annualised against circulating supply. Today the network mints 450 BTC per day, worth $29.63M at the live $65,837.03 spot, an annualised inflation rate of 0.8% — below the historical ~1.5% gold mining rate and far below every major fiat. Four halvings are behind us; twenty-nine remain before the reward rounds to zero.

Three identities, one schedule. For every day d:

btc(d) = reward(d) × 144
usd(d) = btc(d) × price(d)
inflation(d) = (btc(d) × 365) / supply(d)

The reward is fixed by block height. Every 210,000 blocks — roughly four years at the protocol’s 10-minute target — the per-block reward halves. The schedule is hard-coded in GetBlockSubsidy() and enforced by every full node’s consensus rules. There is no policy committee, no emergency override. The reward halves until it rounds down to zero around the year 2140 — at which point the asymptote at 21 million BTC is reached and monetary expansion ends.

144 blocks per day is a target cadence, not a guarantee — and this is the one place the “deterministic” story has slack. Block times follow a Poisson process with a 10-minute mean; difficulty re-centres only every 2,016 blocks. Over a typical month the realised cadence sits in a 140–148 block band, so daily issuance runs a few percent above or below the nominal figure depending on how fast hashrate is arriving.

Supply is the cumulative count of all reward outputs ever issued. It grows monotonically; the first 99% of the eventual 21 million is mined before 2036, with the final 1% stretching across the following century. The full derivation — including the choice of a calendar versus block-height projection for the “days until halving” figure — lives on the methodology page. The point of the three identities is the division of labour: only btc(d) and inflation(d) are monetary policy. usd(d) is a market quantity wearing a policy chart’s clothes.

Reading the series at every halving plus the cycle tops and lows in between is the cleanest chronology of how the schedule has shaped network economics. The BTC column is fully deterministic — it depends only on which halving era the anchor sits in. The USD column moves with the spot price on the date; the inflation column reflects supply growth at that point. The pattern to watch is that the BTC step never moves but the inflation reading keeps grinding lower even between halvings, because supply keeps growing under it.

The distinctive number this chart produces is not the headline inflation reading on any given day — that figure is well-trodden — but the rate at which inflation decays in the year following a halving. Every halving cuts the numerator (daily issuance) in half overnight; the denominator (circulating supply) keeps growing through the new era. The result is a slow further bleed in the inflation rate, layered on top of the overnight cut.

The size of that further bleed has collapsed with each successive halving, because the supply base it divides into has grown. The 2012 halving took inflation from 12.5% to 11.1% over the following year — a 1.4-percentage-point additional decline. The 2024 halving took it from 0.8% to 0.8% over its following year — a 0.01-percentage-point additional decline, two orders of magnitude smaller. The annual decay is now essentially noise.

The single most useful cross-section of this chart is the USD-revenue response across halvings. Each halving cuts BTC issuance in half; the industry’s dollar revenue then depends on whether spot has risen by the same factor. Through the first three halvings the answer was “by far more than that, eventually” — the post-halving year typically delivered a 3–10× price multiplier into the cycle top, more than offsetting the 50% subsidy cut.

The 2024 halving has been a more compressed test. At the cut, USD issuance dropped from $55.20M per day (the 6.25-era closing rate at the day-before close) to $28.56M per day at the new 3.125-equivalent rate — in BTC terms an exact 50% cut, as every halving is, but the largest first-day drop in absolute dollar terms. Today’s flow at $29.63M per day annualises to $10.81B in subsidy revenue across the entire industry, before transaction fees.

The miner-economics frame is the only one where the USD panel adds signal beyond the BTC panel. For monetary-policy questions — how much new supply enters the float, what the real inflation rate is, where the schedule is heading — the BTC panel is the answer. For mining-industry questions, the USD panel is.

If you hold for the long run, the chart is the entire monetary case in one frame: a known, declining issuance schedule with a hard asymptote, executed by every full node without exception. Read the BTC step function and the inflation strip, and ignore the dollar overlay — it is the least informative of the three here because it is dominated by spot, not by the schedule. Pair it with stock-to-flow for the supply-side multiple and the halving cycles overlay for the historical price response.

If you mine, the USD panel is the read. Subsidy revenue alone, today, is $10.81B per year across the entire industry; that figure scales with spot and steps down at every halving. The complement is the daily-fees chart — the second leg of miner revenue, currently small in proportion but projected to dominate post-2140. Subsidy compression has historically been offset by price appreciation; the next halving is the next test of that pattern.

The dollar panel is not the policy. It is price times quantity, dominated entirely by spot. The cleanest demonstration is the 2022 bear: BTC/day never moved off 900 the entire year, yet USD issuance collapsed roughly fourfold from the November-2021 top into the post-FTX low on 2022-11-21 — a pure price move with zero schedule change. A reader watching only the dollar line would have inferred a tightening of supply that never happened. Treat the BTC and USD panels as different questions sharing an x-axis.

The countdown drifts, and so does daily issuance. The “days until next halving” figure is a calendar projection; the actual schedule is block-based (every 210,000 blocks). When hashrate falls hard, blocks slow and the halving arrives later in calendar time — and daily issuance temporarily runs below the nominal figure. The May 2021 China mining ban is the textbook case: hashrate fell more than 50% over the following two months and block intervals stretched well past ten minutes until the July 2021 difficulty adjustment (the largest downward adjustment in Bitcoin’s history) re-centred the cadence. For that window the real BTC/day ran several percent under the schedule.

Subsidy is only one half of miner revenue, and the missing half spikes. Transaction fees are the second leg, usually a single-digit percentage in steady state but occasionally dominant: during the Ordinals/Inscriptions surge in early 2023 and at the 2024-04-20 Runes launch — activated in the fourth-halving block itself (block 840,000) — fees briefly exceeded the subsidy itself. None of that is in the USD panel here, which counts issuance only. For the full revenue picture, read the daily-fees companion. The post-2140 economy lives entirely on that column.

The supply asymptote is not literal zero, and 21 million is rounded high. The integer-satoshi rounding of the reward means daily issuance reaches zero a few halvings before calendar 2140; the 33rd halving rounds the per-block reward to zero satoshis, after which no new BTC is issued. And the 21-million figure is itself a slight overstatement — the asymptotic series totals 20,999,999.9769 BTC if every block were mined exactly on schedule with the coinbase claimed in full, which it has not always been (the 2010 era saw blocks claim less than the full subsidy, permanently lowering the true ceiling).