What is the Merge?

The September 2022 upgrade in which Ethereum stopped using mining and adopted proof-of-stake, joining the existing beacon chain to the transaction layer.

Nothing about balances, addresses, or contracts changed; what changed was who produces blocks and what they are paid, and issuance fell sharply because validators are cheaper to reward than miners were (source: the Ethereum Foundation's Merge documentation). It also ended uncle block rewards and fixed block times at one per slot. Staking withdrawals were not enabled until a later upgrade, which is why deposits made before the Merge were locked for a period afterwards. Mining hardware had no role afterwards, and the proof-of-work chains some miners continued are separate assets rather than continuations of Ethereum. See beacon chain and proof of stake.

It is worth knowing about mainly because of what it did not do, which is most of what people expected.

How it works

Before the upgrade, Ethereum ran two chains in parallel. The original execution chain held all balances and contracts and was secured by mining. The beacon chain, live since December 2020, ran proof-of-stake consensus with validators but processed no user transactions.

The Merge joined them. Consensus moved to the beacon chain, and the execution chain became the payload that beacon-chain validators propose and attest to. No state was migrated, because the execution chain's state simply continued under new management.

Three consequences followed and are often misremembered.

Issuance fell sharply, because validators require far less reward than miners competing on hardware and electricity. The Ethereum Foundation described the reduction as roughly ninety percent (source: the Ethereum Foundation's Merge documentation).

Fees did not fall. Transaction fees are set by demand for block space, which the change did not touch. Anyone expecting cheaper transactions was disappointed, and the fee reduction that did arrive came later from EIP-4844 and rollups.

Withdrawals were not enabled. Staked balances remained locked until the Shanghai and Capella upgrades in April 2023, which is why the beacon chain deposit contract accumulated for over two years before anyone could exit.

Example

Illustrative comparison of what a block producer earns. Under mining, the reward came from a block subsidy plus fees, and the subsidy had to cover hardware and electricity for every competitor who found nothing. Under proof of stake, the reward comes from issuance plus the priority fees and any value from ordering, and the cost of participating is the opportunity cost of the stake plus a modest server. Because the cost side collapsed, the reward side could too, without weakening the security budget in proportion. That is the whole economic argument, and it is why issuance fell without the chain becoming easier to attack.

Why it matters when you buy

The upgrade is the reason Ethereum can be staked at all, which is what the yield section covers. The yield pages show which venues offer staking on an asset and where it is permitted, the chain pages show measured throughput and settlement, and Ethereum covers the asset itself.

Related terms

• beacon chain — the consensus layer that took over
• proof of stake — the mechanism adopted
• proof of work — the mechanism retired
• validator — who produces blocks now
• uncle block — the reward the change ended
• slot — the fixed cadence that replaced mining variance

Questions

Did the Merge make Ethereum cheaper to use?

No. Fees are set by demand for block space and were unaffected. The significant fee reduction came later, from rollups and from EIP-4844 blob data. See blob transaction.

Did I need to do anything as a holder?

Nothing. Balances, addresses, and contracts were untouched. Any message telling holders to migrate or claim was fraudulent.

Is the proof-of-work chain still running?

Some miners continued a separate proof-of-work chain, which is a distinct asset with a distinct market rather than a continuation of Ethereum. See hard fork.