Ethereum’s $2,650 September floor: five myths about Pectra-era supply

The deflationary myth

The belief that Ethereum remains in a permanent state of deflation fails under 2026 conditions. While the "ultrasound money" meme suggested that the EIP-1559 mechanism would always keep the burn rate above issuance, the March 2024 Dencun upgrade changed the math by making Layer 2 transactions much cheaper on the mainnet. I see that since April 2024, the supply grew again. The total supply of approximately 120.7 million ETH no longer shrinks as it did during 2023. The transition to proof-of-stake in September 2022 reduced annual issuance by roughly 88 percent. Before The Merge, the total annualized issuance rate was approximately 4.61 percent. This included a 4.09 percent inflation rate from the execution layer and a 0.52 percent inflation rate from the consensus layer. Prior to the transition, execution layer issuance was estimated at 2.08 ETH per 13.3 seconds, which created a 4.09 percent inflation rate. Consensus layer issuance, based on 14 million total ETH staked, resulted in about 620,500 ETH being issued in a year, or a 0.52 percent inflation rate. Miners previously received 2 ETH per block plus rewards for ommer blocks, totaling about 13,000 ETH per day.

Pectra and validator management

The Pectra upgrade, which went live in May 2025, changes how validators manage their assets. This upgrade allows a single validator to hold up to 2,048 ETH, whereas previously the limit was 32 ETH. This enables large stakers to consolidate their holdings into fewer nodes, which reduces operational complexity. Consensus-layer rewards now compound automatically into the validator balance. You should know that this process removes the need to manually create new validators to maintain stake weight. BitGo supports these workflows and sets an internal maximum of 1,920 ETH per validator to allow for reward accumulation. During consolidation, the source validator stops earning rewards for about 27 hours. This upgrade also reduces the initial slashing penalty by a factor of 128, moving it from 1/32 to 1/4096 of the validator balance. Since the April 2023 Shanghai/Capella upgrade, stakers can withdraw their earnings, and the minimum order amount for subsequent entries is 1 ETH. To ensure stability, the protocol limits the number of validators that can exit, with only about 0.33 percent of the total count able to leave in a given day.

Issuance vs the burn

The gap between daily issuance and the daily burn defines the current supply expansion. In 2026, the protocol issues approximately 2,600 new ETH to validators every day, yet on the quietest days, the network destroys only 50 to 70 ETH through the base fee burn. This massive difference contradicts the idea that the network remains naturally deflationary. To offset 1,700 ETH of daily issuance, the network needs an average gas price of 16 gwei. Because there are 7,200 blocks per day, each targeting 15 million gas, the burn relies on high demand. The relationship follows the function $f(X) = X/108$, where $X$ is the daily issuance. If daily issuance rises to 1,800 ETH, the required gas price to offset it rises to 17 gwei. While the supply can shrink during massive activity spikes, the current reality involves steady growth. Will the network return to sustained deflation if Layer 2 activity continues to dominate the fee landscape? I find the massive disparity between the 2,600 ETH issued and the 50 ETH burned too significant to overlook.