"Deflationary" gets attached to Ethereum the way "digital gold" gets attached to Bitcoin — as a slogan that's partially true in a narrow context, then repeated until the mechanism behind it disappears.
After The Merge in September 2022 and the introduction of EIP-1559 the year before, Ethereum's supply dynamics changed materially. But whether those changes make Ethereum deflationary isn't a yes-or-no question. The honest answer is: sometimes, depending on what the network is doing.
Understanding the actual mechanism matters. It changes what "ETH is deflationary" means, when it's true, and what would have to change for it to stop being true.
How Ethereum's Supply Actually Works
There are two sides to ETH's supply equation: issuance (new ETH created) and burns (ETH permanently destroyed). Whether the net supply grows or shrinks depends on which side is larger.
Issuance
Ethereum still creates new ETH. After The Merge, miners were replaced by validators — participants who lock up (stake) ETH as collateral to propose and attest to new blocks. In exchange, they receive newly issued ETH as rewards.
The issuance rate scales with the total amount of ETH staked. As of early 2026, with roughly 30 million ETH staked, annual issuance runs approximately 0.3–0.5% of total supply per year. That's dramatically lower than the pre-Merge era, when proof-of-work mining required issuing around 4.5% annually to compensate miners.
Burns via EIP-1559
EIP-1559, introduced in August 2021, restructured how transaction fees work on Ethereum. Before it, users bid gas prices in an auction, and all fees went to miners. After, fees split into two parts: a base fee (algorithmically set, paid by users, permanently destroyed) and a priority fee (paid directly to validators as a tip).
The base fee is burned. Not sent anywhere — removed from supply entirely. The amount burned is directly proportional to how much the network is being used. High activity, high base fees, more burned. Low activity, low base fees, less burned.
Net Supply Change
Combine both sides: new ETH issued to validators minus ETH burned from base fees. When network usage is high, burns can exceed issuance — supply shrinks, making Ethereum deflationary during that period. When network usage is low, issuance exceeds burns — supply grows, making Ethereum inflationary during that period.
This isn't a design flaw. It's the actual mechanism, and it means "is Ethereum deflationary?" has a conditional answer: it depends on gas consumption.
Where the Constraints Live
The burn rate is entirely a function of base fee volume, which is a function of demand for Ethereum blockspace. During the 2021 NFT boom and 2024 bull market activity, Ethereum ran deflationary for extended stretches. During quieter periods — particularly as Layer 2 rollups absorbed more activity off the main chain — issuance outpaced burns and supply grew slightly.
There's a Layer 2 effect worth noting here. Rollups (Arbitrum, Optimism, Base, zkSync) process transactions off the Ethereum base layer, then settle proofs or data on-chain. This reduces direct L1 gas consumption. More L2 activity means less L1 base fee burning, which pushes supply dynamics toward inflationary, all else equal.
EIP-4844 (proto-danksharding, implemented in March 2024) specifically added cheap blob space for L2 data — which further reduced L1 gas consumption by rollups. Whether this represents a problem is interpretive. L2 growth is intended — it's the scaling strategy. But it does mean the deflationary supply narrative gets more complicated as L2 adoption grows.
What's Changing
Post-Merge, Ethereum's supply trajectory has been roughly flat with periods of mild deflation and mild inflation depending on activity cycles. The ultrasound.money dashboard tracks this in near real-time — cumulative burns since EIP-1559 have exceeded cumulative post-Merge issuance in net, but the lead fluctuates with activity and isn't permanent.
The blob fee market introduced by EIP-4844 created a separate, lower-cost mechanism for L2s to post data to Ethereum. L2s now pay blob fees rather than expensive calldata costs, and those blob fees feed into their own fee market. The impact on L1 deflationary dynamics is ongoing and not yet settled.
For now: ETH supply is neither strongly inflationary nor strongly deflationary. It's activity-dependent, and the threshold between the two is real but not fixed.
What Would Confirm the Deflationary Direction
Sustained periods where burns exceed issuance across multiple activity cycles — not just bull market peaks. Blob fee burn rates becoming significant relative to calldata burns. Staking issuance remaining stable while L1 gas demand grows. Net supply data showing a consistent negative trend across quarters rather than cyclical oscillation.
What Would Break It
Ethereum would trend persistently inflationary if network activity migrates substantially to L2s without sufficient L1 settlement demand to drive burns, or if staking participation grows significantly, increasing issuance. A protocol change reducing the burn mechanism's scope would also shift dynamics.
"Persistent inflation" on Ethereum at current rates isn't catastrophic — the issuance rate is a fraction of what it was pre-Merge — but it does matter for the supply narrative if that narrative is load-bearing.
Timing
Now: Ethereum's supply is near flat, marginally deflationary in high-activity periods, marginally inflationary in quiet ones. The mechanism is stable and functioning as designed.
Next: EIP-4844's full blob fee market and the impact of growing L2 adoption on L1 gas demand are worth watching over the next 12–18 months.
Later: Future Ethereum upgrades — full danksharding, Verkle trees — will continue to reshape the gas economics, and their supply implications are still working out.
Boundary Statement
This post explains the supply mechanism. It doesn't constitute a claim that ETH is or will be deflationary as a general fact, and it's not an investment argument. Supply dynamics are one variable among many. The active tracking of burn rates, staking data, and supply trends lives elsewhere.
