The economics of Ethereum’s Pectra-era blob fee market

Capacity gains and collapsing fees

The Pectra upgrade on May 7, 2025, changed Ethereum’s data availability economics by increasing the target blob count to 6 per block via EIP-7691. I see this change as a success for L2 scalability because the increased capacity to 8.15GB of data daily reduced the scarcity rollups face. Since the upgrade, rollups buy 20.8% more blobs daily, yet the average number of blobs per block remains 33% below the new target. This imbalance makes blobs "virtually free" for the first time since mid-April 2025. Rollups paid just $0.0000395 in total blob object costs in the five full days following the activation. This is a near 100% decline in blob object fees paid to Ethereum compared to the 60 days before the upgrade. Rollups paid a daily average of $16,250 in blob fees during the 60 days leading into the upgrade. The amount of ETH burned from rollup data posting through blobs declined 71% after activation; it fell from 11.22 ETH daily to 3.26 ETH daily. EIP-4844 originally set the minimum blob base fee at 1 wei, which proved problematic when blob space was underutilized throughout 2024. The Fusaka upgrade in December 2025 addressed this with EIP-7918 by replacing the static 1-wei floor with a dynamic minimum tied to L1 execution costs.

L2 sequencer margins and profitability

L2 sequencer margins improved significantly as data costs fell, which benefits networks with high transaction volumes. I recommend Base for anyone looking at the most efficient economic model because its lack of a native token means all sequencer revenue goes to Coinbase. Base has achieved a 98.54% profit margin after onchain costs, while Linea holds 98.86%. Blast saw its percent margins expand from the high 50% range to over 80% today. The reduction in blob costs improved the profit margins of rollups in both relative and absolute terms; Base became the biggest beneficiary in terms of net income after onchain costs despite transaction costs on some top rollups remaining slightly higher. While Base sees the highest transaction volume, Arbitrum One still generates the largest absolute revenue. You should note that Base’s transaction count exceeds Arbitrum’s on a consistent basis through 2026. Arbitrum generates the largest absolute revenue of any Ethereum L2 due to its high transaction volume and established DeFi ecosystem. Its sequencer margin remains operationally profitable, though its governance has not yet implemented a fee-sharing mechanism to route revenue to ARB stakers. Optimism Mainnet’s sequencer revenue remains lower than Arbitrum’s, as users move to the OP Stack ecosystem. zkSync Era faces different economic pressures because it must pay for cryptographic proof generation, a cost that optimistic rollups like Base and Arbitrum do not incur.

Storage burdens for consensus nodes

Higher blob capacity places a heavier storage burden on consensus layer nodes. These nodes must hold rollup blob data for at least 18 days before they can prune it from their machines. As of May 12, 2025, consensus layer nodes held an estimated all-time high of 44.6GB of unpruned rollup data. If demand sustains, nodes will need to carry an estimated 60GB of data. I find the growing storage requirement for validators to be the most problematic side effect of the Pectra expansion. The increase in blobs purchased daily pushed the amount of data that consensus layer nodes must maintain to a new high. This follows the expansion of the target number of blobs from 3 to 6 and the maximum from 6 to 9. The new target-to-max ratio of 2:3 creates asymmetric fee adjustments. When blocks carry zero blobs, the base fee decreases by approximately 14.5%, while full blocks of 9 blobs increase the fee by only about 8.2%. If the target rate of 6 blobs per block is reached, nodes will have to carry between 95GB and 100GB of data. Can the network maintain stability if storage needs reach the projected 100GB?