Halvings, Burns, and Supply Events
Every few weeks, crypto Twitter erupts over a supply-change headline. A Bitcoin halving is "inevitably" bullish. An Ethereum burn day pushes ETH into "ultrasound money" territory. A token unlock is "about to dump the chart."
This part of Reading Crypto News offers the alternative: a small taxonomy that sorts every crypto supply events headline into one of three buckets, a way to measure magnitude, and a checklist for the stock-to-flow reflex that pretends scarcity alone drives price. Anchored in 2026-current data — Bitcoin's 20 million coin milestone, Ethereum's mildly-inflationary post-Dencun reality, and the ~$6B unlock wave that hit the market in March — this should let you read a supply headline like an analyst rather than a tourist.
The three classes of supply events
Before anything else, fix the vocabulary. Three supply numbers matter:
- Circulating supply — coins liquid and on-market right now, excluding locked, vested, or provably burned coins.
- Total supply — every coin that exists (including locked and vested), excluding coins that have been burned.
- Max supply — the hard cap on how many coins will ever exist. Bitcoin's is 21,000,000. Many tokens have no max.
Nearly every supply headline falls into one of three classes.
Class 1 — Issuance cuts. Scheduled reductions in newly-minted coins. Bitcoin halvings, Litecoin halvings, and changes to proof-of-stake issuance curves are all issuance cuts. They don't destroy coins that already exist; they slow the creation of new ones.
Class 2 — Deflationary burns. Coins actively removed from circulation. Ethereum's EIP-1559 base-fee burn, BNB's quarterly auto-burn, exchange-token buyback-and-burn programs, and one-shot founder burns all belong here.
Class 3 — Timed unlocks. Pre-scheduled releases of previously-locked tokens into circulation. Team cliffs, VC vesting streams, airdrop claim windows, DAO treasury unlocks. These don't change total supply — they move tokens from the "locked" bucket to the "circulating" bucket.
A single project can run all three at once. Ethereum mints new ETH to stakers (issuance), burns base fees on every transaction (burn), and every ERC-20 on top of it has its own unlock schedule. The distinction between native coins vs tokens usually decides whether a story is Class 1/2 or Class 3.
Bitcoin halvings, and the 20 million milestone
The cleanest example of a Class 1 event is the Bitcoin halving. Every 210,000 blocks — roughly every four years — the block subsidy paid to miners is cut in half. The schedule has been 50 → 25 → 12.5 → 6.25 → 3.125 BTC per block. The most recent halving fired on April 19–20, 2024, at block 840,000, reducing the subsidy from 6.25 to 3.125 BTC. That 3.125 figure is the current subsidy and will remain so until the next halving in 2028.
The 21 million cap isn't a policy; it's the sum of a geometric series, and the last satoshi is expected to be mined around the year 2140.
The 2026 anchor is concrete. Bitcoin's circulating supply crossed 20,000,000 BTC at block 939,999 on March 9–10, 2026 — roughly 95.24% of the cap. The remaining ~1 million BTC (about 4.76%) will drip out over ~114 more years. Put differently: 95% of Bitcoin that will ever exist already exists.
Future halvings still matter — they still cut miner revenue in half overnight — but their impact on circulating supply shrinks with each iteration. After the 2024 halving, miner subsidy revenue dropped 50% instantly, fee income became a larger share of miner pay, some miners capitulated, and hashrate dipped briefly before recovering. That's the Bitcoin halving schedule playing out as designed.
How to read halving news: in theory it is fully priced-in — everyone has known the date for a decade — so post-halving price moves reflect narrative flows, not a supply shock. The informative signal is miner behavior and the structure of the block-space market and fees as subsidy fades and fees have to carry security.
Burns: EIP-1559, BNB auto-burn, and buybacks
Class 2 is where the "deflationary" marketing lives. Three examples cover most of the territory.
EIP-1559 activated on August 5, 2021 as part of the London hard fork (block 12,965,000). Every Ethereum transaction pays a "base fee" that is burned outright; only the priority tip goes to the validator. After the Merge on September 15, 2022, ETH issuance to validators dropped roughly 90%, making net issuance (new ETH minus ETH burned) the metric worth watching. For stretches of 2022–2023, net issuance was negative, and the "ultrasound money" framing stuck.
The 2026 reality is quieter. Per ultrasound.money, after the Dencun upgrade in March 2024 shifted user activity to Layer 2s, on-chain base fees collapsed. Burn rates that had run thousands of ETH per day in 2022 cratered to around 50–70 ETH/day in Q1 2025, while issuance stayed near 1,700 ETH/day. ETH has been mildly inflationary at roughly 0.2% per year ever since. Total supply sits near 120.7–121.5M ETH as of April 2026. The burn still exists; it just isn't winning the net-issuance fight.
BNB auto-burn uses a different design. Per BNB Chain's burn reporting, quarterly auto-burns follow a price- and block-driven formula instead of discretionary decisions. The 34th quarterly auto-burn (announced January 15, 2026, the first of 2026) removed 1,371,803.77 BNB, bringing circulating supply to roughly 136.36M. Separately, BEP-95 burns a fraction of each block's gas in real time — on the order of 280,000+ BNB cumulatively since activation. Two independent burn streams, both deterministic.
Buyback-and-burn is the third pattern, common in exchange tokens and some DeFi protocols: route revenue into buying the token on the open market and destroying it. Economically, it's a dividend paid in scarcity rather than cash.
The common failure mode across all three: treating any burn as automatically bullish. A burn only moves the needle if it is large relative to issuance and demand stays stable. Burning 0.1% of supply per year while minting 2% is still net inflationary by 1.9%. Compute net issuance before declaring deflation.
Unlocks and vesting cliffs: the supply event nobody tweets about
Class 3 is where retail most often gets blindsided, because the news cycle rarely covers unlocks until after the chart has already moved.
Two shapes to know. A vesting cliff releases a chunk of tokens on a single date — e.g. "1-year cliff, then monthly vest." On cliff day, a slug of previously-locked tokens becomes transferable at once. Linear vesting drips tokens continuously over a defined period, producing a steady supply stream rather than a one-day shock.
Who is unlocking matters as much as how much. Team, early VCs, advisors, ecosystem wallets, and DAO treasuries all have different sell-pressure profiles. VCs typically sell to return capital to limited partners. Treasuries may deploy tokens into grants or liquidity provisioning rather than selling at market.
Aggregating data from tokenomist.ai and CryptoRank, March 2026 saw roughly $6 billion in previously-locked tokens unlock across ~144 projects — roughly three times the monthly average. Notable 2026 schedules: Sui's community reserve unlock in early March (~43.35M SUI, ~1.13% of circulating supply); Arbitrum's DAO treasury unlock on May 16, 2026 (partially mitigated by governance lag, since treasury tokens don't hit markets immediately); and Aptos' community unlock on May 12, 2026 (~11.31M APT, ~2.59% of circulating supply; the community allocation generally streams linearly rather than landing as a single cliff).
To spot one before it hits your portfolio: check tokenomist.ai, Token Unlocks, CryptoRank's vesting pages, or the project's own tokenomics docs. For any upcoming event, capture three numbers — (1) percent of current circulating supply unlocking, (2) cliff versus linear, and (3) recipient category. A widely-cited historical pattern: roughly 90% of large token unlocks coincide with negative short-term price pressure. Not a trading rule, but a very good reason not to be surprised by one.
How to read supply-event news without the hype
Tie all of this together with a short checklist, plus one warning about the model most frequently invoked in supply-event coverage.
The stock-to-flow trap. Stock-to-flow (S2F) is the ratio of existing stock to annual new flow; higher S2F means more scarcity per unit of new supply. The model became famous for the claim that S2F predicts Bitcoin's price. Empirically, it has missed badly — Bitcoin's 2022 drawdown, the muted response to the 2024 halving, and the general decoupling of BTC price from S2F since 2021 all argue against it as a forecasting tool. Use S2F to compare monetary designs across assets, not to forecast prices. Supply is one input; demand, liquidity conditions, and narrative regimes dominate most of the time.
Circulating versus total. Always ask which number a headline uses. "50% of supply burned" is meaningless if half of supply was locked in a team wallet that nobody was going to sell.
Magnitude test. Compare the supply change to daily trading volume. A 1 million-token unlock into a market doing 500 million per day is noise; into a market doing 2 million per day it's a tsunami.
Timing. Is the event scheduled (public, known for years) or a surprise (emergency burn, unscheduled unlock, policy change)? Markets price in schedules. Surprises move them more.
Put together, the reader's checklist is five questions. (1) Which class — issuance, burn, or unlock? (2) Does it affect circulating or total supply? (3) What percent of current circulating supply is involved? (4) Scheduled or surprise? (5) What is net issuance after accounting for simultaneous burns and unlocks?
Answer those five and the supply-event headline tells you what it actually means. The price chart can do what it wants.
Sources
- The Block and multiple crypto news outlets — reporting on Bitcoin's 20M supply milestone (block 939,999, March 9–10, 2026) and March 2026 token unlock aggregate totals (~$6B, ~144 projects).
- ultrasound.money — Ethereum issuance, burn, and net-issuance dashboards; post-Dencun burn collapse data.
- BNB Chain blog (34th burn announcement, Jan 15, 2026) and bnbburn.info — 34th quarterly auto-burn (Q1 2026), BEP-95 cumulative burn totals.
- Bitcoin Core documentation and block explorers — halving schedule, block subsidy history, April 19, 2024 halving at block 840,000.
- tokenomist.ai and CryptoRank — vesting schedules for Sui, Arbitrum, and Aptos cited in March–May 2026 unlocks.
