The Merge Explained: How Ethereum Switched to Proof of Stake

The Merge was the September 2022 upgrade in which Ethereum's execution layer (the chain that processes transactions and smart contracts) joined with a separate proof-of-stake consensus layer, replacing energy-intensive mining with validator staking as the network's security mechanism. It was one of the most significant changes in the history of a major blockchain, and it happened without a single transaction being lost or a hard restart of the network's history.

Before the Merge, Ethereum used proof of work, the same mechanism Bitcoin still uses. Miners competed to solve computational puzzles using specialized hardware, burning large amounts of electricity in the process. Whoever found a valid solution first got to propose the next block and collect the block reward. This system secured the network, but it consumed roughly as much electricity as a mid-sized country.

What Actually Changed

The Merge replaced that mining competition with proof of stake. Instead of competing with computational power, participants called validators lock up ETH as collateral and take turns proposing and attesting to blocks. A validator that behaves honestly earns rewards; one that acts maliciously or goes offline for extended periods can lose part of its stake through a penalty called slashing.

Technically, the Merge combined two chains that had been running in parallel since December 2020:

• The execution layer (formerly "Eth1"), which processes transactions, runs smart contracts, and manages account balances — this is the part of Ethereum most people interact with.
• The consensus layer (the Beacon Chain, formerly "Eth2"), which had been running proof-of-stake consensus independently, without processing any real transactions, purely to prove the mechanism worked.

On Merge day, the execution layer stopped producing its own blocks through mining and instead began taking instructions from the consensus layer about which blocks to build and finalize. Read more about how this two-part design works in our piece on execution layer vs. consensus layer.

The Energy Impact

The most widely cited effect of the Merge was an estimated 99.95% drop in Ethereum's energy consumption. Proof of work's security comes from the cost of electricity and hardware; proof of stake's security comes from the economic value of staked ETH that can be slashed. Removing mining removed the need for warehouses full of specialized processors running continuously, which is why the energy footprint fell so dramatically.

What Didn't Change

It's easy to overstate what the Merge affected. Several things stayed exactly the same:

• Gas fees. The Merge did not reduce transaction costs. Fee levels are driven by network demand and block space, which the Merge didn't directly address. Scaling for lower fees mostly comes through layer 2 rollups.
• Smart contracts and dapps. Existing contracts, token balances, and decentralized applications continued operating without interruption or code changes.
• The 21 million-style scarcity narrative. Ethereum doesn't have a fixed supply cap like Bitcoin; issuance dynamics changed because validator rewards are lower than mining rewards were, but this is a separate topic from the consensus mechanism itself.
• Withdrawals. Staked ETH couldn't be withdrawn at the time of the Merge — that capability arrived later with the Shanghai/Capella upgrade, covered in our guide to Ethereum withdrawal mechanics.

Proof of Work vs. Proof of Stake, Post-Merge

Aspect	Before the Merge (PoW)	After the Merge (PoS)
Security mechanism	Computational competition (mining)	Staked ETH as collateral
Energy use	Very high	Roughly 99.95% lower
Hardware needed	ASICs/GPUs	Standard server hardware
Entry cost	Mining rig + electricity	32 ETH deposit (or pooled alternatives)
Penalty for misbehavior	Wasted electricity, no direct penalty	Slashing of staked ETH
Block producer selection	First valid proof-of-work solution	Randomly assigned validator

Why the Transition Mattered Beyond Energy

Proof of stake also changes who can participate in securing the network. Mining required significant capital for hardware and ongoing electricity contracts, favoring large industrial operations. Staking, by contrast, lets anyone with 32 ETH run a validator from a home internet connection, and people with less capital can join staking pools. This is explored further in our comparison of solo staking vs. pooled staking.

The shift also changed Ethereum's security assumptions. Under proof of work, an attacker needed to acquire more hashing power than the rest of the network combined. Under proof of stake, an attacker would need to acquire and stake enough ETH to control the network — and because misbehavior is punished by slashing, an attack is directly financially self-destructive in a way that simply outspending honest miners was not.

Risks and Open Questions

The Merge didn't eliminate all centralization concerns. A small number of staking services and liquid staking providers control a meaningful share of validators, which raises questions about client and infrastructure diversity — a topic covered in our article on client diversity. Validators also depend on running reliable software; bugs in dominant client software remain a theoretical systemic risk.

Bottom Line

The Merge replaced Ethereum's energy-intensive mining with a staking-based consensus mechanism, cutting energy use dramatically while keeping the execution layer, smart contracts, and account balances completely intact. It didn't lower gas fees or change ETH's fixed-supply status (it has none), and staked ETH withdrawals only became possible in a later upgrade. If you want to understand Ethereum's current architecture, the Merge is the starting point — but the network's roadmap, discussed in our overview of Ethereum's roadmap, extends well beyond it. Track live network activity and TVL data on MrDeFi's chains page to see how the ecosystem has evolved since.