by Professeur émerite Samuel Furfari, Université Libre de Bruxelles
Abstract
The 2025 edition of the Statistical Review of World Energy marks a significant methodological shift in how global primary energy is measured, particularly with regard to non-combustible renewables such as wind, solar, and hydropower. This change has profound consequences for how the energy transition is perceived, quantified, and narrated politically. This paper summarises the conceptual background, numerical effects and geopolitical implications of this revision, placing it within the broader context of international practice (IEA, Eurostat, UN and EIA). Ultimately, it demonstrates that the previous SRWE methodology significantly overstated the contribution of renewables to primary energy for many years, greatly benefiting their advocates rhetorically.
1. Primary Energy Shapes Geopolitics, Final Energy Impacts Our Bills
The Statistical Review of World Energy (SRWE)[1] has been a key reference for many energy experts for decades. BP – originally known as British Petroleum – began compiling statistics on oil reserves, production, and consumption in 1952. The company’s prominent role in energy geopolitics, dating back to its early association with Winston Churchill (who played a key role in establishing the Anglo-Persian Oil Company, which evolved through several name changes to become BP), has helped establish the BP Statistical Review of World Energy as a leading source for energy geopoliticians.
For over seventy years, this statistical compilation has been a valuable tool for analysing energy trends. Over time, it has expanded to include data on critical minerals and the prices of various energies resources.ed to include data on critical minerals and prices of various consumable energy resources.
In 2022, BP transferred responsibility for maintaining and publishing this historical, comprehensive data collection to the Energy Institute, thereby ensuring its continuation as a vital energy sector resource.
A key concept in energy is the distinction between ‘primary energy’ and ‘final energy’. Primary energy is found in natural energy sources such as coal, oil, gas, uranium, wind, biomass and solar power. Final energy is the energy used by the end consumer.
While nature provides primary energy, it is rarely in a form that is easy to use or handle directly. Converting primary energy into final energy always involves transformation processes that reduce the original energy content. In power plants or oil refineries, for instance, primary energy is converted into more convenient forms, such as electricity or refined fuels. Inevitably, this conversion entails efficiency losses. Consequently, the amount of final energy is always lower than the amount of primary energy, and in some cases, the difference can be significant. Furthermore, final energy must be transported to the end user, causing additional losses, particularly for electricity where energy is dissipated in transmission and distribution lines before reaching consumers.
2. Why Primary Energy Accounting Matters
Primary energy statistics form the basis of almost every high-level discussion on energy, climate and geopolitics. Energy intensity, decarbonisation trajectories, the ‘share’ of different fuels and technologies, and many net-zero scenarios are all based on how primary energy is defined and measured. However, the concept of primary energy in statistics is not based on direct observation; rather, it is the product of conventions.
Two main approaches have coexisted:
- The ‘fossil‑fuel equivalent’ or substitution method, in which non‑combustible electricity (wind, solar, hydro, nuclear, etc.) is expressed as the amount of fossil energy that would have been required in a thermal power plant to produce the same amount of electricity; and
- The ‘physical energy content’ method, in which primary energy for non‑combustible renewables is simply the electricity generated, with no artificial multiplier.
For decades, the SRWE used the substitution method, which made intuitive sense in a world dominated by coal, oil and gas, where almost all electricity was generated by thermal power stations. However, in 2025, in recognition of the growing importance of renewable energy sources and the distortions introduced by this convention, the Review switched to a physical content approach for non-combustible renewables. This aligns the Review with the long-standing practice of the IEA (International Enegy Agency ); Eurostat, the UN system and, more recently, the US EIA (Energy Information Administration).
This methodological shift caused an apparently ‘sudden’ drop in the share of renewables in primary energy and a corresponding increase in the share of fossil fuels, even though there was no dramatic change in the physical energy system from one year to the next.
3. Historical Methodology: The Substitution Paradigm
Using the traditional substitution method, a non-combustible electricity source is given a primary energy content equivalent to the amount of fuel that would have been needed in a standard thermal power station. If the reference efficiency is 38–40%, then 1 kWh of electricity is considered to require approximately 2.5–2.6 kWh of primary energy. In joule terms, 1 kWh (3.6 MJ) of renewable electricity is recorded as approximately 9–10 MJ of primary energy.
This logic was originally intended to enable a consistent comparison of fossil and non-fossil electricity in a system where thermally produced power was standard. It also implicitly highlights the ‘avoided fuel’ that such technologies deliver. While non-combustible renewables remained marginal, the distortion this created in the total primary energy balance was modest.
However, over time, as wind and solar power grew rapidly, the substitution method began to inflate their share of primary energy, producing two major artefacts.
- Overstated renewable share: Renewable electricity was valued at the hypothetical fuel input it displaced, not at its physical output. This significantly boosted its reported share of primary energy.
- Artificial decline in primary energy demand: As systems shifted from fossil-based electricity (with large thermal losses) to renewable electricity (with minimal upstream losses), statistical primary energy demand tended to flatten or decline, even when final energy use was stable or increasing.
In other words, the old method made the system look both cleaner and more ‘efficient’ than it actually was in terms of delivered energy services. It also invited confusion between genuine energy efficiency improvements and mere changes in accounting.
4. The New Method in SRWE 2025
The 2025 SRWE replaces the substitution method for non-combustible renewables with an approach based on physical energy content. Under this framework:
- For wind, solar PV and hydro, primary energy is defined as the electricity generated. In other words, 1 kWh of renewable electricity equals 1 kWh of primary energy (i.e. 3.6 MJ).
- The focus shifts from ‘primary energy consumption’ to total Energy Supply, which is defined as production plus imports minus exports and stock changes, in line with broader international usage.
This transition mirrors the approach taken by the IEA and Eurostat, who have long defined the first usable form of energy from non-combustible renewables as electricity.
It is important to note that nuclear energy does not follow the same shift. For nuclear energy, the SRWE continues to use a thermal equivalent approach, which is consistent with the practice of the IEA and Eurostat. Nuclear primary energy is defined as electricity output divided by a standard efficiency of about 33%. Therefore, 1 kWh of nuclear electricity is equivalent to approximately 10.9 MJ of primary energy.
Similarly, conventional efficiency factors continue to be used for geothermal and certain forms of biomass used for electricity generation, rather than a strict one-to-one mapping between electricity and primary energy. This reflects the underlying thermodynamic conversion.
Therefore, the key change in 2025 concerns non-combustible renewables (wind, solar and hydroelectric power) rather than nuclear power, the statistical treatment of which remains structurally continuous.
5. Numerical Impacts on Global Energy Balances
Using the previous method and the most recently reported global aggregates, the primary energy share for 2024 (published in June 2025) is as follows:
The pattern is consistent: the proportion of non-combustible renewables is roughly halved, while the proportion of fossil fuels increases by several percentage points purely due to the accounting change. The most striking result for non-specialists is the apparent increase in fossil fuels’ share of primary energy.
Before the revision, fossils accounted for around 81.5% of global primary energy. After the revision: fossil fuels account for around 86.7% of the total energy supply.
In absolute terms, however, global fossil fuel use does not suddenly surge. Non-combustible renewables are no longer credited with the hypothetical fuel that would have been burned in their absence; they are credited only with the electricity they physically deliver.
The statistical ‘progress’ previously claimed in terms of a declining fossil share is therefore revealed to have been partly an artefact of the substitution method. The new picture more accurately reflects the underlying physical reality: the global system remains overwhelmingly fossil-based.
6. Alignment with International Statistical Practice
The methodological shift in SRWE 2025 brings it into line with the approaches used by the IEA, Eurostat and the UN Statistical Division:
• For non‑combustible renewables, these organisations treat electricity generation as the primary energy content, with a primary energy factor of one for wind, solar and hydro.
• For nuclear, they use a thermal efficiency factor (often 33%).
• For geothermal and some biomass electricity, they use conventionally agreed efficiencies based on typical conversion chains.
This harmonisation is important. For years, inconsistencies between SRWE and the IEA or Eurostat made global comparisons and cross-reporting difficult. Analysts had to carefully adjust or rebase time series when combining SRWE data with data from other sources. The 2025 edition significantly reduces this issue and makes the SRWE more useful for serious comparative analysis.
7. Implications for Energy Policy and Geopolitics
From a policy standpoint, the new method has two major implications. The fact that renewables’ share of primary energy has halved, from around 14–15% to roughly 6%, shows how heavily global energy systems still depend on fossil fuels, despite the rapid growth of renewable energy capacity and generation. Using this fair methodology, wind accounts for around 2% and solar for around 1%. The renewable energies promoted by the European Commission under the Green Deal represent just 3% of primary energy demand. It should be noted that the promotion of new renewable energy sources is not a consequence of recent climate policies; the development of wind turbines and solar panels began in response to the oil crises of the 1970s. Therefore, it has taken half a century to reach 3%, making 100% target appear unattainable for many decades.
Energy statistics are not neutral in geopolitical discourse. Emerging economies, industrialised countries, and regional blocs have long used ‘shares’ of clean energy in primary supply to support leadership claims or justify burden-sharing. The methodological change in 2025 pushes all actors towards a more objective and consistent set of facts, reducing the scope for selective statistical interpretation.
For many renewable energy advocates, the previous convention provided a convenient backdrop. Charts showed renewables approaching or exceeding 15% of primary energy, creating the impression of rapid structural change. In reality, however, their contribution to final energy services remained far more modest. This distinction was often blurred in speeches, reports and policy documents.
Politically and analytically, the new framework provides a more realistic view of the state of the energy transition. It reveals that the global system is still predominantly fossil-based and that the path to deep decarbonisation is longer and steeper than narratives based on the substitution method suggest. For years, the old methodology allowed the narrative of the rapid ascendancy of renewables to be amplified statistically; the 2025 revision removes that amplification and leaves only the physical data.
As a previous paper by Science-Climat-Energie, ‘Adding energy, not transition: fossil fuels remain the foundation of progress‘, highlighted, what is unfolding is not an energy transition, but an energy addition. Although wind and solar energy are expanding globally, they still account for only around 3% of total primary energy demand. Over the past decade, the absolute increase in fossil fuel consumption has been more than seven times greater than the combined growth of wind and solar energy. Therefore, the gap between fossil fuels and new renewable energy sources is widening, not narrowing.
From an academic perspective, the lesson is clear: energy statistics are not merely technical artefacts; they also shape expectations and policy discourse. Rigorous energy policy must be grounded in metrics that reflect physical reality rather than conventions that inadvertently amplify preferred narratives. In this regard, the 2025 Statistical Review of World Energy is not just a technical update; it is a necessary reality check and a significant step towards intellectual honesty in global energy accounting. It reminds policymakers that their ambitions must be measured against what the energy system can actually deliver, rather than what graphs once seemed to promise.
[1] https://www.energyinst.org/statistical-review