EIP-1559 and the Ethereum Burn: How It Works in 2026

Before EIP-1559, Ethereum used a first-price auction for transaction fees. Users bid an amount, and the highest bidders got included first, which made fees hard to predict and often inflated during busy periods. The proposal, numbered 1559 in the Ethereum improvement proposal process, replaced most of that auction with a protocol-set base fee that rises when blocks are more than half full and falls when they are emptier.

The change went live with the London upgrade in August 2021. Users still add an optional tip to speed up their transactions, but the base fee is no longer a bid to miners or validators. Instead, it is a price set by the protocol according to recent demand, and wallets estimate it for you.

The most discussed part of the change is that the base fee is destroyed rather than paid out. This is the burn. Because the amount burned depends on network activity, the burn rises in busy periods and falls in quiet ones. Understanding the burn starts with understanding that the base fee is a protocol rule, not a payment to a block producer.

Each block has a target size, and the base fee moves toward a level that keeps blocks near that target. If a block is fuller than the target, the base fee rises for the next block. If it is emptier, the base fee falls. The adjustment is limited to a small percentage per block, so fees change gradually rather than jumping.

When a user sends a transaction, the wallet proposes a maximum fee and a tip. The network charges the current base fee multiplied by the gas the transaction uses. The base fee portion is removed from circulation. Any part of the maximum fee above the base fee plus tip is refunded. The tip goes to the block producer.

In practical terms, the steps are as follows. First, the protocol reads the base fee from the parent block. Second, the transaction is included if its maximum fee covers the base fee. Third, the base fee is burned and the tip is paid to the validator. Fourth, the base fee for the next block is recalculated from how full the current block was.

Because the burn is tied to real usage, it is a measure of demand for block space, not a fixed schedule. This is the key difference from a set token burn that happens on a timetable.

Activation came with the London upgrade in early August 2021. Reporting from the first days described more than 5,000 ETH burned in roughly two days, with a value of around 14 million dollars at the time. That is the figure to remember for the very first stretch, because the burn had only just begun and fee activity varied from hour to hour.

In the weeks that followed, the burn accelerated. Later reporting from late August described around 147,000 ETH burned over about 26 days, a value of roughly half a billion dollars at prices at that time. The gap between the early and later figures shows how quickly the burn rose once users and applications adjusted to the new fee model and activity increased.

The first-week story is therefore a growth story. The burn began small and rose with demand. Any figure you read about the first week should be checked against the time period and unit it uses, because a dollar value can look large even when the ETH amount is modest, and the reverse is also possible when prices move.

Past reporting is useful for context, but it is not a live dashboard. For current numbers, use a data source that reports burn in ETH with clear date ranges.

Headlines about large burn totals often mix three units: the number of ETH burned, the dollar value of that ETH, and the cumulative total since 2021. A figure such as 100 million could describe a dollar amount, a cumulative amount, or a mistake in the unit. None of these is the same as the amount burned in one week.

The reporting available for this topic describes first-week burns in the thousands of ETH and early-period burns of a few hundred million dollars over longer windows, depending on the date range. A first-week total of 100 million tokens does not match those descriptions. If you see that claim, look for the original source, the unit, and the exact period it covers.

Another common confusion is treating cumulative burn as weekly burn. Cumulative totals keep rising for as long as the mechanism operates and fees are paid, so a headline about a milestone may describe years of activity. Dividing the milestone by the number of weeks, or checking the daily series, gives a more accurate picture.

When you cite a figure, state the unit, the window, and the source. That habit prevents most errors in burn coverage.

Burning reduces supply, but Ethereum also issues new ETH to validators who secure the network. Issuance depends on how much ETH is staked and on the network's protocol rules. Net supply is therefore the difference between issuance and burn over a period.

After the Merge, the move from proof of work to proof of stake, issuance dropped sharply compared with the earlier mining era. That change made it possible for burn to exceed issuance during busy periods, which is when net supply falls. In quieter periods, issuance can exceed burn, and net supply rises.

This means the phrase deflationary can be true for some windows and false for others. Anyone using the word should specify the period. A week of high activity can show net supply falling, while a week of low activity can show it rising. Looking at burn alone misses the issuance side.

For a clear picture, compare the two series over the same window. Many data tools show burn, issuance and net change together, which makes the relationship easier to read.

Later changes added burn sources beyond the base fee of ordinary transactions. The most important is the blob fee introduced with EIP-4844, which activated in March 2024. Blobs carry data posted by layer-2 rollups to Ethereum, and blob fees use their own market. Blob fees are also burned, which links the health of rollups to the supply picture.

This matters for readers because much of Ethereum's activity now happens on layer-2 networks that post data back to the base chain. When those networks are busy, blob demand can rise and add to burn. When they are quiet, the contribution is smaller. The base chain fee market and the blob fee market are separate, each with its own base fee.

For a beginner, the practical point is that the burn is not only about transactions you send on Ethereum mainnet. Activity elsewhere in the ecosystem can feed the burn indirectly. Checking whether a data source separates execution fees from blob fees helps you understand what drives a given period.

The exact mechanics of blob pricing are more technical than the base fee, so read the protocol documentation if you need precise parameters.

Several kinds of tools track burn and supply. Dashboards from Ethereum-focused data providers show daily burn in ETH, cumulative burn, issuance and net change. General analytics sites such as DefiLlama aggregate fees and activity across many chains, which can help you compare Ethereum with other networks, though you should confirm how each site defines fees and burn.

When you check a figure, follow a short routine. Confirm the unit (ETH or USD). Confirm the window (day, week or cumulative). Confirm whether the figure includes blob fees. Then compare two sources if the number matters for a decision or a publication.

Be careful with live pages, because the numbers update constantly. Note the time you checked, and avoid quoting a snapshot as if it were permanent. Archived pages and dated reports are better for historical questions.

For ETH staking data, protocol pages from liquid staking providers such as Lido and Rocket Pool show their own validator and reward statistics. Those figures describe staking, not the burn, so keep the two topics separate when you read them.

Whether you hold ETH, stake it or only follow the network, the burn does not promise a price move or a fixed return. Burn is one input among many, including market demand, regulation, competing networks and user adoption. Price is set by trading, not by a protocol mechanism alone.

For stakers, returns depend on validator participation, network fees, issuance and any fees charged by the staking provider. Liquid staking adds smart contract risk, depeg risk for derivative tokens and the risk that the protocol itself has a bug or governance failure. Read the protocol's documentation and risk disclosures before you commit funds.

For users who send transactions, the burn is simply part of the fee you pay. Timing transactions for quieter periods, or using wallets that estimate fees, can reduce cost, but the fee rules are the same for everyone.

This is general education, not financial advice. Check the rules that apply where you live, and consider speaking with a licensed adviser if you are making a significant decision.

A common mistake is to think the burn removes fees from the network entirely. The base fee is destroyed, but the tip goes to validators, and validators also receive issuance. The burn is one leg of a two-way system.

Another mistake is to assume the burn is fixed. It moves with demand for block space, so a quiet week can burn very little, while a busy week burns much more. Reading one period as the norm can mislead.

A third mistake is mixing units. A dollar figure can rise or fall because the price of ETH changed, even if the amount burned stayed the same. Always check whether a change reflects ETH volume or price.

Finally, people sometimes assume every chain uses the same mechanism. EIP-1559 is specific to Ethereum, and other networks have their own fee models. Comparing figures across chains requires checking how each one defines and records fees.

To understand the burn, start with the protocol documentation and the original proposal text, then read a dated first-week report and a current data view. Keep a note of the unit, window and source for every figure you use. This simple habit makes the topic much clearer and avoids repeating headline errors.

If you are writing about the topic, state the mechanism clearly, separate the first-week figures from cumulative totals, and say where the numbers come from. If you are following the network for personal reasons, check the data occasionally rather than relying on a single headline.

The EIP-1559 story is ultimately about how a blockchain prices scarce block space. The burn is a visible result of that pricing, and it rises and falls with real use. Understanding that link is more useful than memorising any one milestone.