The math of the sudden Ethereum deflation

The 18,000 ETH burned through Uniswap V4 hook deployments ended a long period of inflationary pressure. For much of 2026, Ethereum’s net supply change sat at an annualized +0.16% because the daily issuance of 2,600 ETH exceeded the gross daily burn of 1,240 ETH. This thirty day streak of ultrasound money deflation marks a massive shift. I find the sudden contraction of supply far more significant than the price volatility seen in recent weeks. The current issuance curve, which includes a 1% annual issuance minus a 0.5% destruction rate, struggled to maintain deflation as blob fees stayed low. This issuance design guarantees a 2% tail APR even if 100% of ETH is staked, a fact that incentivizes every rational holder to stake. When 100% of ETH is staked, all holders face equal dilution, and real profits decrease as staking increases. This structural issue is one reason why researchers suggest a "Croissant Issuance" model with a 50% staking soft cap. Bitcoin’s supply grows at 0.83% annually, which is 66% faster than the supply growth seen in ETH during inflationary periods. Bitcoin’s supply grew by 660,000 in the past week, valued at $66 billion, while ETH’s supply remained stable. A $2 trillion asset like Bitcoin is backed by only $10 billion in economic security, creating a security ratio of 200 to 1. This makes it vulnerable to nation-states like Russia, which can easily launch a 51% attack.

How blobs crushed previous burn rates

The contraction in ETH supply follows a period where Layer 2 networks effectively drained the mainnet. Since EIP-4844 activated in March 2024, transaction fees on Base, Arbitrum, and Optimism dropped by as much as 95% within days. This caused Ethereum’s monthly protocol revenue to fall by 60% to 80% compared to 2023. L2 networks like Base and Arbitrum One process many more transactions than the Ethereum mainnet. In the first quarter of 2026, the combined daily transaction count across all tracked Layer 2 networks exceeded Ethereum mainnet transactions by a ratio between 5:1 and 10:1. You already know that sequencers on these networks capture most of the MEV and user fees. This leaves ETH stakers with much less revenue. Stakers saw fee revenues plummet by more than 90% earlier this year. The reduction in fee revenue happened because L2s transitioned from paying full calldata fees to using blobs. Mean blob utilization across the three dominant rollups climbed to 88% of the 3-blob target. This forced the blob base fee to appreciate to a 2.4 gwei-equivalent, which is a 14-month high. With total value locked across all Ethereum Layer 2 networks exceeding $40 billion in 2026, the economic capture at the L2 layer remains a major issue. Base alone, the network launched by Coinbase in August 2023, has on many days processed more transactions in 24 hours than the Ethereum mainnet.

Tight supply and institutional moves

Supply levels reached lows as users moved ETH into staking and long-term storage. Centralized exchange balances fell to 8.7% last Thursday. This level marks the lowest since the network launched in 2015. Since early July, these exchange balances dropped 43%. This reduction occurred because digital asset treasury purchasing increased. Institutional investors like BitMine and K3 Capital used the recent price declines as a strategic buying opportunity. While retail interest waned due to five years of sideways markets, whales hit all-time highs in accumulation wallets. The ETH/BTC ratio sat at 0.0272 in early June, marking its lowest level since April 2020. Even as ETH faced a 22% drawdown from its February peak of $2,540, the tightest supply environment in years developed in the background. Ethereum prices consolidated around $3,050 in recent observations, though the asset previously faced a close of $1,983 in May 2026. The amount of Ether held on exchanges remained at 8.8% on Sunday, according to Glassnode data.

The looming uncertainty of fee markets

The technical roadmap suggests this volatility might become a regular occurrence. The combined Pectra and Glamsterdam upgrade aims for mainnet activation between September and November 2026. These upgrades aim to increase blob throughput and improve validator efficiency. The market usually front-runs these upgrades by 8 to 12 weeks. The EIP-7691 upgrade specifically targets raising the blob target to 6 and the max to 9. This would double the blobspace supply and likely collapse blob base fees back toward the floor. The sudden transition from an inflationary regime where daily issuance of 2,600 ETH eclipsed the daily burn of 1,240 ETH to this new deflationary phase happened because of massive demand for specific protocol hooks. Will the sudden spike in Uniswap V4 activity remain sustained or will it vanish once the congestion subsides?