EIP-1559

What Is EIP-1559?

EIP-1559 is an Ethereum Improvement Proposal that overhauled how transaction fees work on the Ethereum network. Activated on August 5, 2021, as part of the London hard fork (block 12,965,000), it replaced the unpredictable first-price auction model with an algorithmic fee market where the protocol itself sets a base fee that adjusts with demand.

Before EIP-1559, users had to guess what fee to bid for block inclusion. During congestion, this led to dramatic overpayment as wallets competed blindly. EIP-1559 solved this by making the base fee transparent and predictable: users can see the current base fee and know roughly what they'll pay, with the option to add a small tip to incentivize faster inclusion.

The proposal was authored by Vitalik Buterin, Eric Conner, Rick Dudley, Matthew Slipper, Ian Norden, and Abdelhamid Bakhta. First drafted in April 2019, it was refined for over two years before activation.

How It Works

EIP-1559 introduced a new transaction format (Type 2 under EIP-2718) with two fee parameters instead of a single gas price:

• maxFeePerGas: the absolute maximum the sender will pay per unit of gas
• maxPriorityFeePerGas: the maximum tip the sender will pay directly to the block proposer (validator)

The actual fee per gas unit is calculated as the lesser of maxFeePerGas and baseFee + maxPriorityFeePerGas. Any difference between maxFeePerGas and the actual fee is refunded to the sender. The base fee portion is burned (permanently removed from circulation), while only the priority fee goes to the validator.

Base Fee Adjustment Algorithm

The base fee adjusts each block based on how full the previous block was relative to a target of 50% capacity. Ethereum's gas target is 15 million gas per block, with a maximum of 30 million. This elastic block space allows temporary surges while the base fee corrects demand back toward the target.

The adjustment is governed by two constants from the EIP specification:

• BASE_FEE_MAX_CHANGE_DENOMINATOR = 8 (caps adjustment at 12.5% per block)
• ELASTICITY_MULTIPLIER = 2 (maximum block size is 2x the target)

# Base fee adjustment (simplified from the EIP spec)
if gas_used > gas_target:
    base_fee_delta = base_fee * (gas_used - gas_target) / gas_target / 8
    new_base_fee = base_fee + base_fee_delta

if gas_used < gas_target:
    base_fee_delta = base_fee * (gas_target - gas_used) / gas_target / 8
    new_base_fee = base_fee - base_fee_delta

# Example: base fee = 50 gwei, block used 25M gas (target = 15M)
# delta = 50 * (25M - 15M) / 15M / 8 = ~4.17 gwei
# new base fee = ~54.17 gwei

The relationship is linear: an empty block reduces the base fee by 12.5%, a block at exactly 50% keeps it unchanged, and a completely full block (30M gas) increases it by 12.5%. A sustained sequence of full blocks increases the base fee by roughly 10x every 20 blocks (about 4 minutes), creating strong economic pressure against sustained congestion.

Transaction Lifecycle Under EIP-1559

1. The user's wallet reads the current base fee from the latest block header
2. The wallet suggests a maxFeePerGas (typically 2x the current base fee to buffer against increases) and a small maxPriorityFeePerGas (commonly 1 to 3 gwei)
3. The transaction enters the mempool and validators select it based on its priority fee
4. Upon inclusion, the protocol burns baseFee * gasUsed and pays the validator priorityFee * gasUsed
5. Any unused fee allowance (the gap between maxFeePerGas and the actual charge) is refunded to the sender

Comparison with Bitcoin's Fee Model

Bitcoin uses a pure first-price auction for transaction fees. Users bid a fee rate (satoshis per virtual byte), and miners include the highest-paying transactions first. All fees go directly to miners as part of the block reward.

Bitcoin has not adopted a similar mechanism for several reasons. Its fixed 21 million supply cap is considered foundational: burning fees would permanently destroy coins from a finite supply, conflicting with Bitcoin's monetary design. Additionally, as the block subsidy halves over time, transaction fees must eventually fund 100% of network security. Burning any portion would reduce the incentive for miners to secure the chain. Bitcoin's conservative development culture also resists fundamental changes to the fee market. For a deeper exploration of how Bitcoin's fee dynamics work, see the research article on Bitcoin fee market dynamics.

Economic Impact

EIP-1559's fee burn introduced a deflationary mechanism to Ethereum's monetary policy. Since activation, approximately 4.6 million ETH has been burned. The economic effect has varied across three distinct phases:

Pre-Merge (August 2021 to September 2022)

Under proof of work, Ethereum issued roughly 13,000 ETH per day to miners. While EIP-1559 burned significant amounts (over 2.6 million ETH in the first year alone), daily issuance generally exceeded daily burns. The net supply remained inflationary, but at a reduced rate compared to pre-1559.

Post-Merge Deflation (September 2022 to March 2024)

Ethereum's transition to proof of stake (the Merge, September 15, 2022) cut issuance by approximately 90%, from roughly 13,000 ETH per day to about 1,700. Combined with EIP-1559 burns, ETH became net deflationary. The circulating supply decreased by roughly 300,000 ETH in the first year after the Merge.

Post-Dencun (March 2024 Onward)

The Dencun upgrade activated EIP-4844 (blob transactions), creating a separate fee market for Layer 2 data. This dramatically reduced the fees L2 rollups paid to mainnet, causing burn rates to collapse. As of mid-2026, ETH is mildly inflationary at roughly 0.23% annually. The deflationary threshold requires average gas fees above approximately 16 gwei, a level the network has generally remained below since Dencun.

EIP-4844 and Multi-Dimensional Fee Markets

EIP-1559's design has been extended beyond execution gas. EIP-4844 (proto-danksharding), activated in March 2024, introduced a second, independent EIP-1559-style fee market specifically for blob data used by rollups. Blob gas has its own base fee that adjusts independently using the same algorithm, creating a multi-dimensional pricing model where each type of blockspace resource is priced according to its own supply and demand.

Future proposals continue this trend. EIP-7706 proposes a third gas dimension for calldata, which would create three independent EIP-1559-style fee markets operating simultaneously. The core insight: different blockchain resources have different scarcity profiles and benefit from independent pricing mechanisms.

Why It Matters

EIP-1559 is significant beyond Ethereum because it demonstrated that blockchain fee markets can be designed rather than simply auctioned. The predictability improvements changed how wallets estimate fees, how users experience transactions, and how the protocol manages demand.

For projects building across multiple chains, understanding fee market design is essential. Spark operates as a Bitcoin Layer 2 where transactions settle off-chain without per-transaction gas fees, taking a fundamentally different approach to the fee problem that EIP-1559 addresses. While Ethereum chose to make on-chain fees more predictable, Layer 2 solutions like Spark and the Lightning Network move transactions off-chain entirely.

Risks and Considerations

Validator Revenue Reduction

By burning the base fee, EIP-1559 reduced the direct revenue validators earn from transaction fees. Validators only receive the priority tip, which during periods of low demand can be minimal. This has shifted the economic model for validators, making staking rewards (block issuance) a larger share of their income.

MEV Incentives

Reduced tip revenue has amplified the relative importance of maximal extractable value (MEV) for validators. Since EIP-1559 limits what validators earn from standard fees, the economic incentive to extract value through transaction ordering, front-running, and sandwich attacks has become comparatively more important.

Base Fee Manipulation

In theory, validators could attempt to manipulate the base fee by producing artificially empty or full blocks. However, the 12.5% per-block cap limits the impact of any single block, and coordination across multiple validators is difficult to sustain. The mechanism has proven robust in practice over five years of operation.

Burn Volatility

The deflationary effect of the burn depends entirely on network activity. During high-demand periods (NFT mints, DeFi volatility), burns can far exceed issuance. During quiet periods, the burn may be negligible. This creates unpredictable supply dynamics that differ from Bitcoin's deterministic halving schedule.

This glossary entry is for informational purposes only and does not constitute financial or investment advice. Always do your own research before using any protocol or technology.