Gas Fees as Demand Signal: Interpreting Spikes and Lulls

As of March 26, Etherscan's gas tracker shows Ethereum average gas prices at approximately 0.04 Gwei, with basic ERC-20 transfers costing roughly one cent. This is the cheapest Ethereum has ever been to use. Simultaneously, the network recently processed a record 2.88 million daily transactions, the highest count in its history. For an allocator who has been told that gas fees reflect network demand, these two data points appear contradictory: record usage and near-zero fees. Understanding why they are not contradictory, and what gas fees actually signal in 2026, is essential for reading on-chain fundamentals accurately.

What Gas Fees Actually Measure

Gas fees on Ethereum measure the price of block space, not the level of economic activity. Each Ethereum block has a gas limit that caps the total computational work it can include. When demand for block space exceeds supply, users bid up the gas price to get their transactions included. When supply exceeds demand, prices fall toward the minimum.

This is a congestion metric, not an activity metric. The distinction matters because Ethereum's effective supply of block space has expanded dramatically. The Dencun upgrade in March 2024 introduced blob transactions that reduced Layer 2 data posting costs by over 90 percent. Layer 2 networks now process more than 18 times Ethereum's Layer 1 transaction volume, handling the bulk of user-facing activity while settling to Ethereum for finality. The result is that Ethereum L1 block utilization sits around 46 percent on recent days, well below capacity, even as total ecosystem transaction volume reaches all-time highs.

The gas fee metric in 2026 tells you whether L1 block space is congested. It does not tell you whether the Ethereum ecosystem is being used.

How It Gets Confused

Two common misreadings persist. The first treats falling gas fees as evidence of declining demand. In 2021, when Ethereum was the only execution layer and DeFi activity concentrated on L1, gas fees were a reasonable proxy for demand. Average gas prices reached 224 Gwei in February 2021. By March 2025, they had fallen to approximately 2 Gwei, a 99 percent decline. But Ethereum ecosystem transactions, including L2s, grew throughout this period. The fee decline reflects expanded capacity, not reduced usage.

The second misreading treats fee spikes as evidence of healthy growth. During the 2021 NFT boom, gas fees spiked because users paid hundreds of dollars per transaction to mint speculative assets. A gas spike driven by an NFT launch has different economic implications than one driven by institutional settlement. The metric does not distinguish between them.

MEV activity adds another distortion. When arbitrage bots compete for transaction ordering, they bid up gas prices in ways that do not represent end-user demand. Fees can spike 10 to 20 times above normal levels in a single block. These spikes are real expenditures but represent infrastructure extraction, not economic growth.

Companion Metrics

Three companion metrics reduce false signals when reading gas fees. First, total ecosystem transaction count across L1 and all major L2s. Ethereum L1 recently hit 2.88 million daily transactions, but L2 networks collectively process over 14 million transactions per week on L1 alone, with far higher counts on their own execution layers. The total ecosystem throughput is the demand signal. Gas fees are the congestion signal.

Second, ETH burned via EIP-1559. Since August 2021, a portion of every base fee is permanently burned. When burn rate exceeds issuance, ETH becomes deflationary. In 2025-2026, persistent low fees have produced minimal burn, resulting in slight net inflation. The burn rate ties gas activity directly to ETH's monetary dynamics.

Third, weekly base-layer fee revenue in dollar terms. DeFiLlama data shows Ethereum L1 weekly fees averaged roughly 2.3 million dollars over the past month, down from approximately 8 million dollars during a February spike. This metric captures the economic value users actually pay for block space, combining gas price and transaction volume into a single revenue figure that is directly comparable across periods.

What a Healthy Trend Looks Like

Over 12 months, a healthy Ethereum ecosystem shows rising total ecosystem transactions and rising fee revenue in dollar terms, while gas prices in Gwei remain stable or decline modestly. This pattern indicates growing usage absorbed by expanding capacity without congestion. An unhealthy pattern is rising gas prices in Gwei without corresponding growth in unique addresses or transaction diversity, which signals congestion from speculative or bot-driven activity rather than organic demand growth.

The upcoming Glamsterdam upgrade, scheduled for H1 2026, targets raising the gas limit toward 100 million with parallel block verification, projecting up to 78 percent further fee reduction. For a banking examiner, the question is not whether fees are low but whether the network's fee revenue, inclusive of L2 settlement payments, sustainably funds its security budget. Whether L2 growth translates into L1 settlement demand or cannibalizes L1 activity entirely remains the most important open question in Ethereum's economics.

For informational purposes only. Not an offer to buy or sell any security. Available only to accredited investors who meet regulatory requirements.

‍