The Ethereum Merge: What Changed After It Went Live

The Merge was the event on September 15, 2022 when Ethereum's main network stopped using proof-of-work mining and began using proof-of-stake. Before the Merge, two separate networks ran side by side: the Ethereum mainnet, which processed transactions and smart contracts, and the Beacon Chain, a proof-of-stake network launched in December 2020 to coordinate staking.

The Merge joined them. The execution layer, which handles transactions, account balances and smart contract code, was attached to the consensus layer, which decides which block is added next. After this, miners were no longer needed to secure the chain. Validators, who lock up ETH as collateral, took over that role.

The event was planned as a transition rather than a restart. Balances, contracts and transaction history carried over. Users did not need to move funds or take any action to keep their ETH, and the network was designed to keep producing blocks throughout the switch.

For readers asking what the Merge was in one line: it replaced mining with staking as the way Ethereum reaches agreement on the state of its ledger, while keeping the same ETH, the same addresses and the same contracts.

Before the Merge, Ethereum was secured by miners who competed with specialised hardware to add blocks. After the Merge, validators run software that proposes and attests to blocks. To become a validator, a participant locks 32 ETH for each validator. Validators that follow the rules earn rewards, and those that break the rules or go offline can lose part of their stake.

The design split Ethereum into two layers. The execution layer runs the EVM, processes transactions and stores state. The consensus layer runs proof-of-stake, chooses blocks and finalises them. Each validator runs a consensus client and an execution client alongside it. Most post-Merge discussion about node operators, client diversity and staking infrastructure comes from this split.

For a non-technical reader, the practical change is that a validator's security depends on their stake and their software setup, not on electricity and hardware. Slashing rules exist to punish provable double signing or conflicting attestations, and inactivity penalties apply when a validator goes offline for long periods.

When you read about Ethereum's architecture now, remember that the layer separation is the Merge's main structural legacy. Later upgrades have mostly added features to these two layers rather than replacing them.

The most visible change was energy use. Proof-of-work required miners to run large amounts of hardware continuously. Because proof-of-stake does not need that competition, Ethereum's energy use fell by a very large share. Widely cited estimates put the reduction at around 99.9 percent, and that figure is the one most commonly repeated.

The reduction did not come from making the blockchain faster or cheaper to use. It came from removing the mining process. Proof-of-stake still requires computers to run validator software, and those computers still use power, but the scale is far smaller than mining rigs.

Miners were the other side of the change. Their income ended on Ethereum, and many turned to other proof-of-work networks or sold their equipment. The change also meant GPU supply and power contracts that had supported Ethereum mining lost that demand.

If you see claims about Ethereum's environmental impact in 2026, ask which period they describe. Figures for the proof-of-work era and the proof-of-stake era are not comparable, and the post-Merge figure is the one that applies to today's network.

After the Merge, the amount of new ETH issued to validators became much lower than the issuance under proof-of-work mining. Public analysts reported a large cut in annual issuance, and that is the change most often described as making ETH less inflationary. The exact percentage depends on how many validators are active and how the calculation is done, so treat any single number as a snapshot.

Issuance is only half of supply. Ethereum also burns part of each transaction fee under the fee mechanism introduced in the London upgrade in August 2021. When network activity is high, more ETH can be burned than issued, and net supply can fall. When activity is low, net supply can still grow.

The Merge therefore changed the baseline for supply, while the fee burn decides the net result from one period to the next. Readers should avoid assuming ETH is always deflationary or always inflationary.

To check current issuance and burn, use a block explorer or a public dashboard that tracks ETH supply, and note the date. Compare the figures over several weeks rather than one day, since activity changes quickly.

Several rumours circulated before the Merge, including claims about downtime, a fork that would split the network, and fees that would drop sharply. The Merge was designed to avoid a chain halt. The switch happened at a planned point, and the network continued producing blocks. Users did not need to upgrade their wallets to keep their balances.

Fees did not fall because of the Merge. Gas prices still move with demand. Ethereum's main chain remains a relatively expensive place for small transactions at busy times, and many users moved activity to layer-2 rollups, which post batches of transactions back to Ethereum. The big reductions in everyday fees have come from that layer-2 model and later data upgrades, not from the proof-of-stake switch.

Some early articles promised faster transactions straight after the Merge. Block times and finality are governed by the consensus rules, and the change to proof-of-stake did not make individual transactions settle faster for users in a noticeable way.

If you want to know current fees, check the gas price on a block explorer or in your wallet before you transact, and compare with a layer-2 network if your transaction is small.

Staking is the core activity after the Merge. Anyone can run a validator by locking 32 ETH per validator and operating the required software on reliable hardware with a stable connection. Running your own validator gives full control, but it also means you carry the operational responsibility. Mistakes such as keeping keys on a poorly secured machine, or running duplicate validators, can lead to penalties.

For smaller holders, staking pools let people contribute less than 32 ETH and share rewards. Pool design differs, so read the terms, fee structure and withdrawal rules before depositing.

Withdrawals were not enabled at the Merge. Validators could stake, but they could not withdraw their stake until a later upgrade. The Shapella upgrade in April 2023 enabled withdrawals, so stakers gained a way to move their staked ETH out of the consensus layer. Before that date, some stakers had to wait with their ETH locked.

When you consider staking now, check the current validator queue and exit times on a block explorer, since these depend on network activity. Check the reward rate on the platform you use, and note that rates change with total staked ETH.

Liquid staking protocols let users stake ETH and receive a token that represents their staked position. The token can often be used in other DeFi applications, which is why liquid staking became a large part of the post-Merge ecosystem. Lido and Rocket Pool are two of the best-known examples. Lido issues stETH, and Rocket Pool issues rETH, and both protocols describe their operator model and governance on their own sites.

Liquid staking adds risks on top of ordinary staking. The token can trade at a discount to ETH when markets are under stress, the protocol's smart contracts can contain bugs, and the operator set can concentrate risk. Governance decisions can change fees or rules, so read the current documentation for each protocol before you rely on its token.

Liquid staking tokens can also be used as collateral or inside lending markets. That creates useful options for some users, but it also means a problem in one protocol can spread to another. Keep track of where your tokens are used.

For a current view of total staked ETH by protocol, use a public staking dashboard and check the date. Market share changes over time, and any percentage in an article becomes out of date.

The Merge was not the last upgrade. Shapella in April 2023 enabled withdrawals. The Dencun upgrade in March 2024 introduced blob data for layer-2 rollups, which lowered the cost of posting their data to Ethereum and helped reduce fees on many rollups. Later upgrades in 2025 continued the work on validator tools and account features, and the Ethereum Foundation's public roadmap describes the direction, including scaling through rollups, data availability improvements and further validator changes.

The pattern since the Merge is clear. Ethereum keeps its base layer focused on security and settlement, and it pushes scaling and everyday transaction volume toward rollups and other networks. Readers who follow the roadmap should check each upgrade's status on official channels, because timelines have moved in the past.

For users, the practical result is that the Merge is one stage in a longer process. Asking whether something is 'Ethereum' now means asking which layer, which rollup, and which upgrade you are discussing.

When you read about future upgrades, check whether they are proposals, testnet releases or live mainnet activations. Those three stages are different, and articles often blur them.

A useful way to review the Merge today is to separate what happened from what is still changing. What happened: mining ended, validators took over block production, energy use dropped sharply, and ETH issuance fell. What is still changing: fees, layer-2 usage, staking rates, validator counts and the upgrade schedule.

If you hold ETH, the Merge itself does not require any action. Check your wallet's supported networks, make sure you have a secure backup of your seed phrase, and keep your software current. If you plan to stake, decide between running a validator, using a pool, or using a liquid staking protocol, and read the documentation of the one you pick.

If you are reading about the Merge for research or work, use primary sources where possible. The Ethereum.org documentation, protocol research posts and upgrade announcements are more reliable than summaries that repeat old predictions.

Finally, treat every number as dated. Staked ETH, validator counts, gas prices and issuance change constantly. Record the date and source whenever you use one, and recheck before you act on it.