Fee Burn

A fee burn destroys a slice of every transaction fee by rule. Ethereum's EIP-1559, live since August 2021, burns the base fee of each transaction, and Tokenomist calls it the cleanest burn in crypto for a specific reason: it is automatic and funded purely by network demand, with no discretion and no treasury behind it. Nobody funds it, and nobody has to remember to execute it. Other chains have copied the shape with different dials. Monad burns 100% of base fees under an EIP-1559 model, permanently and without redistribution, while priority tips still go to the validator, and it charges the base fee on the gas limit rather than on gas used. Aptos burns most of a transaction fee: execution and I/O gas is destroyed permanently, while the storage portion is burned and re-minted when a refund is due, which is why Aptos itself now describes fees as mostly burned. Solana burns half of the base fee only, 5,000 lamports per signature, and since SIMD-0096 activated at epoch 741 on 12 February 2025 priority fees go entirely to the validator with nothing burned, so most of the SOL spent on fees survives.

Because the burn is a function of usage, the interesting number is not how much has been destroyed but where the deflation threshold sits: the activity level at which burned fees start to exceed new issuance. The issuance side of that comparison is the knowable half. Monad pays 18 MON per block on 300 millisecond blocks after MIP-12 activated at round 89,758,000 on 23 July 2026, about 1.89 billion MON a year or roughly 1.9% of the 100 billion initial supply as of 14 August 2026, and its base fee burn has to clear that run rate before net supply contracts. Below the threshold the protocol is still inflating; above it, supply contracts. Solana shows how wide the gap can be: signature fee burn runs at roughly 650 SOL a day against roughly 60,000 SOL a day of inflation, so even with the inflation schedule stepping down from 8% to 1.5%, fee burn is nowhere near offsetting issuance.

Crossing the threshold is not a one-way door. Ethereum had burned roughly 4.63 million ETH cumulatively as of mid-2026, but after the Dencun upgrade pushed activity onto Layer 2s, mainnet fees and therefore the burn fell below staking issuance, and total supply has been growing again since, running roughly 0.2% to 0.8% a year net as of August 2026 with supply up about 950,000 ETH since the Merge. The burn is real; it is simply not winning, which is the clearest available proof that a burn does not automatically shrink supply. The single most important test for any burn is whether it outpaces new issuance, and a burn headline means nothing until you net it against emission. Governance can also move the goalposts: Aptos burns most of every transaction fee today, with no validator or treasury share, but on-chain governance can vote to change that policy.

A fee burn is one cell in Tokenomist's two-dimensional burn taxonomy, not a synonym for it. By Type it is Programmatic, firing without human intervention; by Reason it is Protocol Design, because token destruction is part of how the chain operates. That combination is the most dependable deflationary force available, and it is worth contrasting with a scheduled burn on the same token. BNB runs both: BEP-95 burns a governance-set share of the gas fees collected by validators in real time on every block, 10% when it went live on 30 November 2021 and revisable by validator proposal and vote, with the remainder passed to the ValidatorSet contract for validators and delegators. The quarterly Auto-Burn is scheduled destruction rather than usage-driven, sized from BNB's average price, BNB Chain block production and a governance-set constant, with those inputs adjusted after the Lorentz, Maxwell and Fermi upgrades shortened block times. Its 36th round on 15 July 2026 destroyed 1,615,827.795 BNB and left supply at 133.17 million. Only the first is a fee burn. This activity dependence is also why Tokenomist's future emission projections include only inflation from known cliff and linear unlock schedules and deliberately exclude burns, since burn rates are unpredictable and protocol-dependent. Historical emission counts every burn in full; the forward number stays conservative rather than overstating deflation that may never arrive.