Ethereum’s transition from proof-of-work to proof-of-stake — “the Merge” — on 15 September 2022 was the most significant technical upgrade in blockchain history. More than two years later, the data is in. The verdict: a resounding success on energy, a mixed picture on decentralisation, and a continuing challenge with staking concentration.
The energy impact has been dramatic. The Merge reduced Ethereum’s energy consumption by approximately 99.95%, according to research from the Cambridge Centre for Alternative Finance. A single Ethereum transaction, which previously consumed as much energy as an average US household uses in a week, now uses roughly the same as a few Google searches. This fundamentally changed the narrative around blockchain’s environmental impact and removed a major obstacle to institutional adoption.
Staking participation has exceeded expectations. As of March 2025, approximately 33 million ETH — roughly 27% of the total supply — is staked across more than one million validators. The annualised staking yield has ranged from 4% to 6%, depending on network activity and the number of active validators. For comparison, US 10-year Treasury yields over the same period averaged around 4%, making ETH staking yields competitive with traditional fixed income in an environment of elevated interest rates.
The centralisation concern, however, is real. Lido, a liquid staking protocol, controls approximately 30% of all staked ETH — a level of concentration that raises concerns about the network’s resistance to censorship at the protocol level. Lido is a DAO, not a single company, but the concentration of validator nodes in a handful of entities (Coinbase, Binance, Kraken, Lido and a few others) means that the Ethereum network’s censorship resistance relies on the independence of a relatively small number of institutional actors. The Ethereum community is aware of the problem and has discussed solutions including validator set caps and further decentralisation of liquid staking. None have been implemented yet.