Cambridge research finds Ethereum rights prove low energy consumption

Ethereum's post-merger energy consumption performance: At a lower level in the proof-of-stake network

Ethereum's operational energy consumption performance after the completion of the merger is cleaner than in its proof-of-work era. But a new assessment by the University of Cambridge reveals a more complex reality: among the major proof-of-stake blockchains, Ethereum has a low energy intensity, but its total power consumption is still higher than most other proof-of-stake networks in this comparison.

A research report released by the Cambridge Center for Alternative Finance estimates that Ethereum consumes approximately 7.87 gigawatt hours of electricity every year. On an economically adjusted basis (i.e., energy consumed per unit of market value), the network corresponds to approximately 33 kilowatt-hours per million dollars of market value, ranking the second lowest among the proof-of-stake networks evaluated, after the BNB chain.

Core Discovery

Based on Cambridge University's estimates of node-level power consumption, Ethereum consumes approximately 7.87 GWh annually. Its energy intensity is approximately 33 kilowatt-hours per million dollars of market value, which is at the lowest level in the measured proof-of-stake network. Among the networks studied, Solana consumed the highest annual electricity consumption, reaching approximately 13.48 GWh. Node location and grid structure have a significant impact: The University of Cambridge believes that the remaining emissions mainly come from the power system that powers Ethereum verification nodes and infrastructure. In Ethereum's energy structure, renewable energy and nuclear energy account for more than fossil fuels, and Cambridge estimates that the proportion is 56.4%.

Ethereum's energy intensity positioning

The Cambridge study is part of an ongoing effort to quantify blockchain sustainability, using measurable input data rather than macro assumptions. The research focuses on the proof-of-stake phase of Ethereum and aims to provide policymakers and investors with a basis for comparison of network energy consumption that is closer to current reality. Using its analytical framework, Cambridge not only estimated overall power consumption, but also established a method to compare energy consumption with the economic scale of each network. Under this adjusted perspective, Ethereum's market value per million US dollars of approximately 33 kilowatt hours is second only to the BNB chain among the proof-of-stake networks evaluated. In contrast, the report showed that Solana consumed the highest electricity in the comparison group, at approximately 13.48 gigawatt hours per year. Cambridge also calculated that Solana's energy intensity is approximately 283 kilowatt-hours per million dollars of market value, which is approximately 8.5 times that of Ethereum, while the total power consumption of all networks in the comparison group is approximately 38 gigawatt hours.

How Cambridge models node energy consumption

The core part of the report is an attempt to map Ethereum's power consumption to actual operating conditions. The University of Cambridge measured "wall-side" power consumption in standard consumer-grade and professional-grade hardware configurations through 20 Ethereum node software client combinations. In the model, a "typical home" configuration consumes approximately 18 watts of power, while a workstation-level environment consumes approximately 153 watts of power. Combining these parameters with the observed distribution of Ethereum node types and hosting models, Cambridge estimates that the average power consumption of a representative node is approximately 105 watts. In order to calculate the total consumption of the entire network from equipment power consumption, approximately 8522 discoverable complete nodes were studied and counted. Cambridge's distribution estimates show that 64% of these nodes operate through cloud services or corporate facilities, with the remaining 36% connected through residential networks. This distinction is crucial because power grids vary significantly in different regions. Cambridge concluded that Ethereum's remaining emissions come mainly from the power grid that powers these nodes-a finding that is important for sustainability discussions because discussions tend to focus on the blockchain protocol itself rather than on its surrounding infrastructure.

Proportion of renewable and nuclear energy in Ethereum's energy structure

In addition to total power consumption, the study also assessed the composition of the energy mix used by Ethereum node operators. Cambridge estimates that 56.4% of Ethereum's power input comes from renewables and nuclear energy, with the remaining 43.6% coming from fossil fuels. This ratio is a key detail for readers concerned about what "low energy intensity" actually means. A network may be relatively efficient by value, but depending on where its infrastructure is hosted and how the local grid provides power, it may still rely partially on fossil energy for electricity generation. In the broader context of sustainability claims, this approach shifts the discussion to operational decisions-such as hosting preferences and regional power structures-rather than viewing energy consumption as a purely intrinsic attribute of the network.

From Proof of Work to Proof of Equity: Reasons for Changes in Numbers

To understand Ethereum's current energy consumption situation, one must review the major changes it has experienced. Ethereum passed a merger in September 2022, transitioning from proof-of-work mining to proof-of-stake verification. This change replaces miners competing through energy-intensive computing and is instead maintained by validators who pledge ether to secure the network. The Cambridge report is also consistent with the widely cited post-merger conclusion that the upgrade significantly reduced power consumption by removing mining. It has been previously reported that the merger reduced power consumption of the Ethereum network by more than 99.9%, and Cambridge's latest assessment further extends this theme by quantifying the remaining energy consumption after the protocol change. In other words, Ethereum's energy intensity is now determined more by validators and infrastructure choices than by consensus-related hardware competition-making current estimates more reflective of modern operating conditions than any forecast in the mining era. Looking ahead, as node hosting models and regional energy structures change, investors and policymakers may need to focus on how these estimates evolve and whether new measurement methods will continue to refine power consumption, market value, and real infrastructure.