Ethereum’s average gas price fell to 5.8 gwei in a post-Dencun low

Ethereum mainnet’s average gas price fell to 5.8 gwei during the UTC day of May 12, 2024, according to Etherscan’s daily series. A report preserving the reading described it as the lowest daily average in the explorer’s dataset, which begins with Ethereum’s 2015 launch period.

The milestone mattered for two reasons. Users were paying unusually little for Ethereum layer-one blockspace, but the same decline reduced the base fees destroyed under Ethereum’s fee-burning mechanism. Cheap execution was therefore both a usability improvement and a warning that mainnet demand—and the burn supporting ether’s supply economics—had weakened.

What the measurement shows

Gas measures the computational work consumed by an Ethereum transaction. Gas price is normally quoted in gwei, with one gwei equal to one-billionth of one ether. Etherscan’s statistic is the average effective gas price paid by transactions included across the UTC calendar day; it is not a wallet quote, a median, or the price paid by every user.

For scale, multiplying the 5.8-gwei daily average by the 21,000 gas units required for a standard ETH transfer produces an illustrative cost of 121,800 gwei, or 0.0001218 ETH. That is a calculation, not an observed universal fee. Smart-contract interactions consume more gas, individual transactions paid different prices, and converting the result into dollars would require a separately defined ETH price and timestamp.

Contemporaneous reporting on May 9, 2024, had already placed gas near 5 gwei at a point in time, down from approximately 63 gwei on March 12. The figures are not directly interchangeable: one is an intraday observation and the May 12 figure is a full-day average. Even so, they establish that the decline was sustained rather than a single-block anomaly.

Dencun changed the setting

Ethereum’s Dencun upgrade activated at epoch 269568 on March 13, 2024, at 13:55 UTC. Its most consequential scaling change, EIP-4844, introduced temporary data blobs intended to reduce the cost of posting rollup data to Ethereum.

Dencun did not simply make every layer-one computation cheaper. It created a separate, lower-cost data market for rollups, strengthening the incentive to execute activity on layer-two networks while using Ethereum for settlement and data availability. Contemporaneous analysis also attributed the low mainnet reading to activity moving toward layer-two systems and to softer demand for layer-one execution.

That distinction limits any causal claim. The 5.8-gwei observation verifies what users paid on May 12; it does not isolate how much of the decline came from Dencun, migration to rollups, market conditions, application activity, or ordinary weekend variation.

Why low gas affected ether’s economics

Under EIP-1559, Ethereum’s base fee rises when blocks exceed their gas target and falls when usage is below that target. The protocol burns the base-fee portion of transaction payments, while validators receive priority fees. Lower gas prices therefore tend to mean less ETH burned for a given amount of gas consumed.

By May 8, contemporaneous reporting based on CryptoQuant and Ultra Sound Money data said issuance had exceeded burning over the preceding 30 days. That did not establish a permanent inflation rate, and the May 12 gas reading alone could not determine supply growth. It did show the mechanism behind the change: affordable blockspace reduces fee destruction unless greater usage offsets the lower price.

Limits and later context

Etherscan’s 5.8-gwei value is a UTC daily arithmetic average and may be influenced by unusually expensive transactions. It should not be presented as a median user experience or proof that all Ethereum transactions were inexpensive.

A May 15 report said the daily average rebounded to roughly 10 gwei on May 13 and May 14. That later observation reinforces the limits of treating one quiet day as a permanent fee regime. The durable conclusion from May 12, 2024, is narrower: Ethereum mainnet blockspace reached an exceptional low-cost point while the network was adapting to a rollup-centered scaling model.