The Ethereum Merge (2022): How Ethereum Moved From Mining to Staking
On 15 September 2022 Ethereum switched from proof of work to proof of stake. See how the Merge worked, what changed, and what it did not change.
On 15 September 2022, Ethereum changed the way it reaches agreement on new blocks. The network stopped using miners and proof of work and began using validators and proof of stake. The upgrade, known as the Merge, was one of the most complex changes ever made to a running blockchain, and it was completed without taking the network offline.
Why change at all
Proof of work, used by Bitcoin and by early Ethereum, secures a network by making block production costly in computing power and electricity. Ethereum's developers had planned from the beginning to move to proof of stake, in which validators lock up cryptocurrency as a deposit and are chosen to propose and approve blocks. If they behave dishonestly, part of the deposit can be destroyed, a penalty known as slashing. The main reasons given for the move were lower energy use, a different security model and a foundation for later scaling work.
The road to the Merge
| Date | Step |
|---|---|
| 1 Dec 2020 | The Beacon Chain, a separate proof-of-stake chain, launches and runs in parallel |
| Aug 2021 | The London upgrade introduces EIP-1559, which burns part of each transaction fee |
| Jun 2022 | Ropsten test network completes a trial merge, followed by other test networks |
| 15 Sep 2022 | Mainnet merges with the Beacon Chain |
| Apr 2023 | The Shanghai and Capella upgrades allow staked ether to be withdrawn |
The Beacon Chain existed for almost two years before the Merge, with validators staking ether on it, though it did not process user transactions. The Merge joined the existing Ethereum execution layer, where accounts, balances and smart contracts live, to the Beacon Chain as the consensus layer. In effect, the old mining system was replaced beneath a running application layer.
How the switch was triggered
The switch was timed according to a value called terminal total difficulty, which measures the total mining work accumulated on the chain. When the chain reached the set threshold, the next block was produced by the new system rather than by miners. This meant the exact time depended on mining power in the days before, and it occurred in the early hours UTC on 15 September. Applications and users did not need to take any action, and contracts kept working.
How proof of stake works on Ethereum
- Validators. Running a validator requires staking at least 32 ETH.
- Slots and epochs. Time is divided into 12-second slots. In each slot, one validator is chosen to propose a block and others attest to it.
- Finality. Blocks are finalised after a couple of epochs, meaning they become extremely costly to reverse.
- Penalties. Validators that are offline lose small rewards; those caught signing conflicting blocks can be slashed.
Energy use
The Ethereum Foundation stated that the Merge cut the network's energy consumption by about 99.95 percent. Independent researchers also estimated a very large reduction, since validators run ordinary computers rather than banks of specialised mining hardware. The exact figure depends on the assumptions used, but the direction is not disputed. The Foundation also estimated a reduction of roughly 90 percent in new ether issuance, because validators are rewarded with much smaller amounts than miners were.
What the Merge did not do
There were claims before the upgrade that it would solve several other problems. It did not:
- Reduce transaction fees. Fees are set by demand for block space, which the Merge left unchanged. Cheaper fees have come mainly from layer-2 rollups.
- Increase transaction throughput directly. Block capacity stayed as it was.
- Allow staking withdrawals immediately. That came with the Shanghai upgrade in 2023.
Client diversity
After the Merge, an Ethereum node runs two pieces of software together: an execution client and a consensus client. Several independent teams build each kind, such as Geth, Nethermind and Besu on the execution side, and Prysm, Lighthouse, Teku and Nimbus on the consensus side. Diversity matters because a bug in a client used by a large majority of validators could affect the whole network. Validators need reliable, always-on machines, which some operators rent as a dedicated server.
Reactions and trade-offs
Most of the Ethereum ecosystem supported the change, but not everyone. A group of miners created a fork that continued proof of work, known as Ethereum PoW, which attracted little activity. Ethereum Classic had kept its own proof-of-work chain since 2016. Some critics argued that proof of stake can concentrate power among large holders and staking services, or raised concerns about the influence of a small number of providers. Supporters point to the penalties and the broad distribution of validators. The debate is ongoing and involves design judgments, not simple facts.
Why it matters beyond Ethereum
The Merge showed that a large, value-bearing network could change its core security mechanism without a shutdown. It also gave regulators and companies a reason to look at energy use differently when assessing crypto. Bitcoin remains on proof of work, and its supporters give their own reasons for that choice, including its security record.
A practical lesson for site owners
The Merge was planned over years, tested on staging networks and rolled out in stages. That is a good model for your own site: test major changes on a copy before applying them live. If you run development environments or blockchain nodes, a VPS plan from Zebra Hosting gives you the control to experiment safely.
