Blobs remain nearly free after Pectra capacity expansion

The Pectra upgrade increased the target blob count to 6 per block and the maximum to 9. This change pushed daily data capacity to 8.15GB, yet rollups purchase 33% less than the new target. Because usage stays well below the target rate, blobs cost rollups less than one-thousandth of a penny daily. This oversupply makes blob space virtually free for the first time since mid-April 2025. I see a market where the supply of data space remains much higher than the current rollup demand. Since Pectra went live, rollups purchase 20.8% more blobs daily than they did before the upgrade. In the 60 days leading into the upgrade, rollups purchased 21,200 blobs per day. In the five full days following the activation, they purchased 25,600 blobs daily. Rollups also purchased 2.7 gigabytes of data capacity daily pre-upgrade against 3.3GB today. The median cost per blob object since Pectra went live remains $0.00000000035. Consequently, rollups pay a near 100% decline in blob object fees compared to the 60 days leading into the upgrade, when rollups paid a daily average of $16,250.

Base benefits and node data load

Base benefits most from this surplus. Between August 6 and September 4, 2026, Base generated $3,259,170 in gross revenue from 284,653 blobs. I find the imbalance between these L2 profits and the cost to the network difficult to ignore. You know how much L2s benefit from these low fees, but I see a widening gap between rollup profits and network maintenance costs. Consensus layer nodes must hold rollup data for at least 18 days before they can prune it. This requirement pushed the estimated unpruned data to 44.6GB. While the Pectra upgrade increased the daily data capacity to about 8.15GB, rollups have not yet reached a level of demand to consistently hit the new target rate of 6 blobs per block. Base accounts for 27.35% of the included volume in the recent sample. If current demand sustains, consensus layer nodes must carry around 60GB of rollup data. If demand hits the target rate, nodes must carry between 95GB and 100GB.

For a fleet of 12,000 nodes, the central network cost reaches roughly $0.5 million per year. However, the direct blob base-fee burn only reached 0.293 ETH per week. This gap shows that current fees do not cover the infrastructure expenses of the node fleet.

Governance and the minimum fee

EIP-7918 introduces a minimum blob fee to address this imbalance. This mechanism ties the floor to execution costs to prevent the market from staying in a near-zero state. This redistribution of value from L2 operators to ETH holders creates a direct conflict between stakeholders. L2 teams like Base and Optimism hold significant influence. They will push back if the protocol tries to raise the floor to cover node costs. The proposed minimum blob base fee of 2^25 wei equals about 1/5th of the originally proposed $0.05 value. This amount targets the cold-start problem. This change might increase costs for efficient submitters by 10% to 30% if they carry five blobs. The current minimum base fee of 1 wei makes the market highly vulnerable to price volatility during demand surges. Because the market lacks a reserve price, it takes much longer for fees to rise during periods of high activity. Will the community accept a higher floor to subsidize node infrastructure?