How primary energy is measured has changed across our charts
The Energy Institute no longer uses the substitution method to estimate primary energy. Here’s why that matters for our charts on energy.
Some of the most popular charts on Our World in Data are about energy production and consumption: how much energy countries produce, how they compare per person, and what sources this energy mix comes from.
But when people talk about “energy”, they’re not always talking about the same thing. There are different ways of measuring energy, from the very top of the chain — for example, coal going into a power plant — to the final link in the chain — the light that eventually comes out of a lightbulb.
The differences between the first stage (primary energy) and the last (useful energy) can be very large, so it’s important to be clear about which metric is being referred to when people speak about data on “energy”.
In a previous article, I walked through the four different ways of measuring energy. I’ve included the main visualization below.
On Our World in Data, we present data on both primary energy and secondary energy in the form of electricity generation. We would like to present final and useful energy too, but international datasets on these metrics are, unfortunately, not openly available.
But even when it comes to primary energy, there are differences in how this is calculated. Recently, the methodology used by our main data source changed. In this article, I want to explain this change and how it affects our charts on this topic.
The challenge of comparing energy sources using primary energy
The main difference in the calculation of primary energy comes down to how we account for the energy wasted when we burn fossil fuels.
If you put coal into a power plant, around two-thirds of its energy value is wasted as heat, and only one-third is converted into electricity. For gas, roughly half is wasted, and half is converted to electricity. So to get 1 kilowatt-hour (kWh) of electricity out, you need to put in the equivalent of 3 kWh of coal and 2 kWh of gas. I’ve sketched this in the diagram below.
This is not the case with non-combustible sources, such as solar, wind, and hydropower: they just deliver electricity. We just see 1 kWh of electricity output; there’s no upstream fuel input or waste along the way.
That means that 1 kWh of electricity from coal, 1 kWh from gas, and 1 kWh from solar will look very different when measured in primary energy terms. It would be equivalent to 3 kWh of coal, 2 kWh of gas, and 1 kWh of solar. Expressed as a share of primary energy, coal would make up half of our energy mix, gas one-third, and solar one-sixth. Even though the electricity we get out — and can actually use — is split evenly between coal, gas, and solar, each makes up one-third.
The old methodology: substitution method
Our main data source for primary energy is the Energy Institute’s Statistical Review of World Energy.1
It historically presented data on primary energy using the substitution method.
This method attempts to account for the inefficiency of fossil fuels by converting renewable and nuclear energy into their “fossil equivalents”. In other words, it asks: if this electricity had been generated from fossil fuels instead, how much fuel would have been used?
To get this figure, it divides renewable and nuclear electricity generation by a typical thermal power plant efficiency factor, which is around 38% to 41%. So, if a country generates 100 terawatt-hours (TWh) of electricity from solar, that amount is divided by 40% to get a “fossil equivalent” of 250 TWh.
This means renewable energy figures are all inflated by a factor of around 2.5.
The net effect of this is that non-fossil fuel sources are measured as the amount of fossil fuel inputs they would have substituted. This has two impacts:
- Total reported primary energy is higher than reality because we have inflated the true figures for non-fossil sources (the 100 TWh of solar power generated is included in the total as 250 TWh).
- Fossil fuels make up a smaller share of the primary energy mix than if they were measured directly. Arguably, this share is a better reflection of their contribution to energy demand, because in relative terms, it strips out some of the inefficiencies. This is still not a perfect reflection of energy demand, because it’s not based on final or useful energy.
The new methodology: physical energy content
The Energy Institute no longer publishes primary energy data based on the substitution method.
Instead, it’s based on the physical energy content method, giving a measure of total energy supply. In our charts, we still refer to this as primary energy.
The physical energy content method does not make any adjustments to solar photovoltaic (PV), wind, and hydropower generation. 100 TWh of solar electricity generation is included as 100 TWh of primary energy (rather than 250 TWh in the substitution method approach). It does still make adjustments for other renewables — geothermal, concentrating solar, and biomass — where heat is an input (these have thermal efficiency factors applied to them).
It still measures fossil fuels in terms of their raw, calorific value — how much energy is produced when burned (including wasted energy).
Nuclear power generation is treated slightly differently. It still uses a thermal equivalent conversion. This is because nuclear plants operate like a thermal plant: fission heat is the input for steam, which turns a turbine to generate electricity. To get the total nuclear energy supply, the electricity output is divided by the average thermal efficiency of nuclear, which is around 33%. So, 100 TWh of nuclear power generation would equate to around 300 TWh of physical energy input.
Almost all of our charts on primary energy now use this physical energy content method in line with the data source; if you have previously followed or used these metrics, you might notice a change in the numbers.
Here’s how that change matters for the numbers:
- Primary energy from solar PV, wind, and hydropower drops significantly, because it’s no longer inflated to a “fossil fuel equivalent”.2
- Primary energy from nuclear increases a bit, because its own thermal efficiency factor is actually lower than was used as a “fossil equivalent” in the previous method. This is also true for geothermal energy.
- In most cases, the share of primary energy coming from fossil fuels increases compared to the previous substitution method, mostly because of the drop in primary energy from solar, wind, and hydropower (the first bullet).
In the chart below, I’ve shown how 100 TWh of coal, solar, and nuclear power compare under the substitution method versus the physical energy content method.
A few final things to note
This is purely an accounting change; nothing physical in the energy system has changed.
It applies to the entire historical series of primary energy, not just the most recent year. That means the methodology applied to the data is consistent across decades.
Endnotes
This was previously published as the BP Statistical Review of World Energy.
This is not the case for some other, smaller renewable energy sources, such as geothermal.
Cite this work
Our articles and data visualizations rely on work from many different people and organizations. When citing this article, please also cite the underlying data sources. This article can be cited as:
Hannah Ritchie and Pablo Rosado (2026) - “How primary energy is measured has changed across our charts” Published online at OurWorldinData.org. Retrieved from: 'https://archive.ourworldindata.org/20260807-163119/primary-energy-measurement-change.html' [Online Resource] (archived on August 7, 2026).BibTeX citation
@article{owid-primary-energy-measurement-change,
author = {Hannah Ritchie and Pablo Rosado},
title = {How primary energy is measured has changed across our charts},
journal = {Our World in Data},
year = {2026},
note = {https://archive.ourworldindata.org/20260807-163119/primary-energy-measurement-change.html}
}
Reuse this work freely
All visualizations, data, and articles produced by Our World in Data are completely open access under the Creative Commons BY license. You have the permission to use, distribute, and reproduce these in any medium, provided the source and authors are credited.
The data produced by third parties and made available by Our World in Data is subject to the license terms from the original third-party authors. We will always indicate the original source of the data in our documentation, so you should always check the license of any such third-party data before use and redistribution.
All of our charts can be embedded in any site.