EIP-1559 is the Ethereum Improvement Proposal that restructured how transaction fees work on Ethereum. It replaced the first-price auction model — where users guessed a gas price and miners picked the highest bidders — with a protocol-set base fee plus an optional priority fee. The proposal activated on August 5, 2021 with the London hard fork at block 12,965,000, and has since burned more than 4.6 million ETH from circulation according to ultrasound.money.
​

The change addressed a long-standing usability problem. Under the legacy auction, users routinely overpaid for gas because they could not see what the next block would clear at. EIP-1559 made the fee predictable for the next block, with a small variance, and gave wallets a cleaner abstraction to display.
​

EIP-1559 is a Core Standards Track EIP authored by Vitalik Buterin, Eric Conner, Rick Dudley, Matthew Slipper, Ian Norden, and Abdelhamid Bakhta. It was first published as a draft in April 2019, refined for over two years, and shipped with the London upgrade. The full EIP-1559 specification on eips.ethereum.org remains the canonical reference.
​

The core mechanism splits each transaction's gas fee into two parts:

Users specify a maxFeePerGas (the ceiling they will pay) and a maxPriorityFeePerGas (the tip cap). The wallet computes min(maxFeePerGas - baseFee, maxPriorityFeePerGas) as the actual tip, and any unused fee is refunded. This pattern eliminates most overpayment.
​

The base fee adjusts per block based on a target of 15 million gas with a hard cap of 30 million gas. If a block fills above target, the base fee rises by up to 12.5% in the next block. If it fills below target, the base fee falls by up to 12.5%. This feedback loop pushes blocks toward 50% capacity on average.
​

The mathematics is deterministic. Given the previous block's gas used and base fee, anyone can compute the next block's base fee exactly:

The 12.5% adjustment cap means a sustained surge of 30M-gas blocks doubles the base fee in roughly six blocks (about 72 seconds at 12-second slots). A sustained drought halves it in the same span. This bounded volatility is what makes the next-block fee estimable.
​

The burn is the second-order effect. Because the base fee is destroyed rather than paid to a validator, every transaction reduces ETH supply. As of mid-2024, cumulative burn exceeded 4.6 million ETH worth roughly $13 billion at then-prevailing prices. The Etherscan burn chart tracks the daily flow.
​

Three problems motivated the change. The first was overpayment. Under the first-price auction, users submitted whatever gas price their wallet suggested, and that suggestion was historically calibrated to clear under congestion. During calm periods users routinely paid 2-5x the market-clearing rate because there was no signal of where the next block would settle.
​

The second was fee volatility. Block-to-block gas prices could swing by 50% or more, especially around NFT mints and arbitrage opportunities. Volatility punished users who set fees once and waited; their transactions either got stuck or overpaid relative to the actual cleared price.
​

The third was MEV (maximal extractable value) leakage to miners. Under the auction model, miners earned both block rewards and the full gas fee, giving them an incentive to manipulate inclusion order or selectively reorder transactions. EIP-1559 removed the base fee from miner revenue, leaving only the tip as a behavioral lever. The change is part of a broader push, alongside Paradigm's research on MEV markets, to keep validator economics aligned with users.
​

The exact formula from the EIP-1559 reference implementation is:
​

base_fee_next = base_fee_prev + base_fee_prev * (gas_used - target) / target / 8
​

The division by 8 caps the per-block adjustment at 12.5%. The target is half the gas limit (15M of 30M). Worked example: a block uses 25M gas, base fee was 50 gwei. Adjustment = 50 * (25M - 15M) / 15M / 8 = 50 * 0.0833 = 4.17 gwei. Next block's base fee = 54.17 gwei.
​

The minimum base fee is 7 wei. Below this, the algorithm cannot adjust further down. In practice, base fees during quiet periods sit at 1-3 gwei rather than approaching the floor.
​

For users, the visible effect is wallets that quote a single estimated fee instead of asking users to pick a gas price. MetaMask, Rabby, Coinbase Wallet, and Phantom all handle the maxFeePerGas / maxPriorityFeePerGas split internally and surface only the dollar estimate. Wait time is still a function of the priority fee, but the base fee is no longer a guess.
​

For validators (the term replaced miners after the September 2022 Merge), revenue from gas now consists only of priority fees and any MEV they capture, plus consensus rewards. Average priority fees during normal traffic sit around 1-2 gwei. During congestion — large NFT mints, liquidations, airdrop claims — priority fees can spike to 50+ gwei.
​

For ETH supply, EIP-1559 turned ETH into a partially deflationary asset. Combined with the post-Merge issuance reduction (from ~13,000 ETH/day under proof-of-work to ~1,700 ETH/day under proof-of-stake), Ethereum has spent extended periods with negative net issuance. The ultrasound.money dashboard visualizes the daily inflation/deflation rate.
​

Consider a quiet weekday morning on Ethereum mainnet. Block 22,500,000 has used 12 million gas (below the 15M target). The base fee for the next block falls 2.5%, from 8 gwei to 7.8 gwei. Users submitting transactions during this window see wallet quotes around 8 gwei plus a 1 gwei priority tip — roughly $0.30 for a simple ERC-20 transfer.
​

By midday, an airdrop claim opens. Demand surges. Block 22,500,300 hits the 30M gas cap. The base fee rises 12.5% to 8.78 gwei. Two blocks later it reaches 11.1 gwei. Three minutes after the surge began, the base fee has roughly doubled. Users who set their maxFeePerGas at the morning's 12 gwei find their transactions stuck because the base fee now exceeds their cap.
​

Power users react by raising maxFeePerGas. The priority fee market also activates — users compete on tip to get included in the next block. The two markets — base fee (predictable, exponential) and priority fee (auction, discrete) — coexist for the rest of the surge.
​

By evening, demand falls back. The base fee decays at 12.5% per under-target block, returning to roughly 5 gwei within an hour of the airdrop closing. Users who waited save substantially compared to those who pushed through during the peak.
​

Several claims about EIP-1559 circulate that the data does not support:

EIP-1559's predictable fee structure simplifies how stablecoin orchestration platforms estimate transaction costs. Cross-chain solvers — including Across, LI.FI, and Eco — quote total fees that include destination-chain gas. The base fee component of that quote is computed from the next-block estimate rather than a worst-case bid, which reduces the spread between quoted and realized cost.
​

EIP-1559 has also been adopted, with variations, by most EVM-equivalent L2s. Arbitrum, Optimism, Base, Polygon zkEVM, and zkSync all implement 1559-style base fee mechanics, though parameters (target gas, max adjustment) differ. This is one reason cross-chain quote engines can apply consistent fee logic across stablecoin developer tools and B2B payout APIs.
​

Eco's intent-based cross-chain routing uses the next-block base fee on each candidate chain to score destination cost. Without EIP-1559, the quote would have to include a wider buffer. With it, quotes can be tighter and still reliably clear.
​

EIP-1559 activated on August 5, 2021 at block 12,965,000, as part of the London hard fork. The proposal had been in development since April 2019. Activation required every Ethereum client — Geth, Nethermind, Besu, OpenEthereum — to ship coordinated updates ahead of the fork block.
​

Cumulative burn exceeded 4.6 million ETH by mid-2024, with the rate varying widely based on network demand. The 2021 NFT boom and 2022 ETHUSD volatility produced months with negative net issuance. Quieter periods, especially after L2 adoption surged in 2024, have brought net issuance positive again.
​

Most EVM-equivalent L2s implement EIP-1559-style fee markets, but with different parameters. Arbitrum uses a target gas of 7 million; Optimism uses a similar approach. The base fee on L2s is generally far smaller because rollups inherit Ethereum security at a fraction of the cost, especially after EIP-4844 reduced data posting fees.
​

The legacy gasPrice field still works for backwards compatibility, but new transactions use Type 2 (EIP-1559) format with maxFeePerGas and maxPriorityFeePerGas. Most wallets default to Type 2. Validators include both legacy and 1559 transactions in the same block.
​

The activation was widely anticipated and partially priced in. ETH rose from roughly $2,500 to $4,800 between August and November 2021, but attribution to EIP-1559 alone is hard to isolate from broader 2021 cycle dynamics. The supply effect is real, but a single fork rarely drives sustained price movement on its own.
​