The World Steel Association, worldsteel, estimates that the steel sector emits 7% – 9% of global CO₂ emissions  But other numbers are available: take your pick of anything from 6%¹ to 10%² or 11%³. Where do all these numbers come from – and what do they mean?

Let’s start with the worldsteel figures.  Here’s their statement, in full:

“In 2020, on average, every tonne of steel produced led to the emission of 1.89 tonnes of CO₂ into the atmosphere. In 2020, 1,860 million tonnes (Mt) of steel were produced, and total direct emissions from our sector were of the order of 2.6 billion tonnes, representing between 7% and 9% of global anthropogenic CO₂ emissions.”⁴

So we have a nice, simple equation: steel sector CO₂ emissions on the top, out of global anthropogenic CO₂ emissions on the bottom, and that should give us 7% to 9%. We’ll look at the top half first.

Steel sector CO₂ emissions

In this case, steel sector emissions refers to the production of crude steel, including stainless steel – but not including emissions from processing, manufacturing, or casting downstream of crude steel production.

The worldsteel statement refers to the direct emission of 2.6 billion tonnes of CO₂. But it also gives an average emission figure of 1.89 tonnes of CO₂ per tonne of production, and a total production volume of 1,860 million tonnes which works out at 3.5 billion tonnes of CO₂ emitted. So in addition to the 2.6 billion tonnes of direct emissions their figures show 900 million tonnes of indirect emissions.

To worldsteel’s credit, the detailed methodology for their determination of CO₂ emissions is published on its website⁵. Their calculation of indirect emissions takes account of emissions from power generation and process gases (allowing credits for their subsequent re-use), but does not include CO₂ or other greenhouse gas emissions associated with mining of the input materials for steelmaking⁶. So they are not counting any of the emissions for the mining of iron ore, metallurgical coal or limestone, or of the metals used for the production of stainless or high alloy steels; and of course methane is excluded (the calculation is for CO₂ emissions only, not for all greenhouse gases).

Global CO₂ emissions

So much for the top half of the equation – what about the bottom half. 

The United Nations Environment Programme (UNEP) publishes annual summaries of data on global emissions⁷, disaggregated for CO₂ and other greenhouse gases and for what is known in the jargon as LULUCF – emissions and removals of greenhouse gases resulting from direct human-induced land use, land-use change and forestry activities.

The UNEP reports show an anomolous decrease of emissions for the year 2020 due to the Covid pandemic. But figures for both 2019 and 2021 are quite similar, with a central estimate for 2019 of 44.3 billion tonnes (Gt), and a range of around +/-10%⁸.

	Steel sector CO2 emissions (Gt)	Global CO2  emissions (Gt)			Steel sector contribution (%)
		low	central	high	low global	central	high global
Direct	2.6	39.9	44.3	48.7	6.5%	5.9%	5.3%
Indirect*	0.9				2.3%	2.1%	1.9%
Total	3.5				8.8%	7.9%	7.2%

Assuming that emissions associated with LULUCF are included in what worldsteel refers to as ‘global anthropogenic CO₂ emissions’, we can now consider steel’s contribution to the global total. That gives a central estimate of 7.9%, and a range from 7.2% (using the high end of the range for global CO₂ emissions) to 8.8% (using the low end of the range for global CO₂ emissions) – in line with the 7% to 9% range given by worldsteel.

What then accounts for the variation between the worldsteel figures and those given in other reports, of up to 11% of global emissions?

One possible source of variation is the year for which a comparison is made. Global CO₂ emissions for 2023, for example, were estimated at 43 billion tonnes – slightly lower than for 2019 (possibly due to a change in the methodology for calculating emissions relating to land use and forestry). At the same time worldsteel’s estimate for the CO₂ emissions intensity of steel production has increased from 1.89 to 1.91 tonnes of CO₂ per tonne of production; and the volume of crude steel production has also increased, from 1,860 to 1,892 million tonnes. Putting those figures together raises the steel sector’s relative contribution.

Another difference is that many authors reference the International Energy Agency’s estimate for indirect emissions of 1.1 Gt, rather than worldsteel’s (2019) value of 0.9 Gt. IEA used a different methodology to determine indirect emissions – but if their figure of 1.1 Gt is combined with the estimated 2.6 Gt of direct emissions, and 2023 global emissions data, then the contribution of the steel sector is increased to between 8% and 10%.

A more concerning variation is the fact that, as noted above, worldsteel does not include the upstream emissions of mining in their calculation of steel sector emissions. They are not the only organisation to discount this contribution, but it seems strange to imply that the emissions associated with the mining of iron ore and metallurgical coal should not be allocated to the production of steel. Including these upstream emissions will increase the steel sector’s contribution to the global total.

As it happens, the CO₂-only emissions associated with mining iron ore and metallurgical coal are low when compared to the emissions of steelmaking itself. It is hard to find CO₂-only data, but a back-of-envelope estimate suggests that these emissions may be less than 50 million tonnes of CO₂ per year. However, the figures are only so low because they exclude methane, which is the biggest factor by far. That will have to be the subject of another post.

If we update the worldsteel data to 2023, and add a conservative estimate for CO₂ emissions associated with mining, we get the figures shown in Table 2.

	Steel sector CO2 emissions (Gt)	Global CO2  emissions (Gt)			Steel sector contribution (%)
		low	central	high	low global	central	high global
Direct & indirect*	3.6	38.7	43.0	47.3	9.3%	8.4%	7.6%
Mining	0.04				0.1%	0.1%	0.1%
Total	3.7				9.4%	8.5%	7.7%

So what does it all mean?

Firstly, when worldsteel says ‘CO₂‘ it really does just mean CO₂. It is not including other greenhouse gases, and in particular methane. If other greenhouse gases were included the absolute value of emissions would be higher, although steel’s contribution as a proportion of global emissions might not be: you can’t simply add on the steel sector’s methane emissions without also including methane in the figure for total global emissions.

Secondly, the figures should include mine-related emissions – it’s absurd that they don’t.

Thirdly, the figures need to be updated. Including the emissions from mining, the sector’s contribution to global CO₂ emissions in 2023 was between 7.7% and 9.4%, with a central value of 8.5%. So the range last year was 8% to 9%, and it may, in 2024, already be 8% to 10%.

The comparison with global emissions is useful in that it gives us some sense of scale. But it’s also a composite value, dependent on three factors: how much steel is produced each year; the average emissions per tonne of steel produced; and global CO₂ emissions. Steel production is projected to increase substantially between now and 2050, and we all hope global CO₂ emissions will come down. Unless steel’s emissions intensity falls faster than the combination of those two factors its relative contribution to the global total is going to increase. Scrutiny of the sector’s performance is going to grow.

But finally, what drives climate change is not steel’s relative contribution to global emissions, but the absolute level of emissions. Steelmaking is emitting 3.7 billion tonnes of CO₂ a year, and that is growing. Including methane will put the figure up to more than 4 billion tonnes of CO₂ equivalent.

Those are huge numbers. Governments – yes, governments – need to be putting more effort into reducing them.

Footnotes

1. Reduction of Greenhouse Gas Emissions in Steel Production – Final Report.  Pandit et al (2020). ↩︎
2. Iron and Steel Technology Roadmap: Towards more sustainable steelmaking, IEA (2020). ↩︎
3. Building a decarbonised steel sector.  Sullivan K and Olson G, World Resources Institute (2023) ↩︎
4. worldsteel website accessed 10 November 2024 ↩︎
5. https://worldsteel.org/climate-action/climate-action-data-collection/ ↩︎
6. Nor does it include emissions for the extraction of coal, oil or natural gas for power generation. In accordance with World Resources Institute guidance for the determination of Scope 2 emission factors, such upstream emissions are excluded. According to WRI they may be included by electricity users within their own Scope 3 Category 3 emissions – although in reality this is completely impracticable (GHG Protocol Scope 2 Guidance, para 4.2). ↩︎
7. United Nations Environment Programme (UNEP) ‘Emissions Gap’ reports, with data from EDGAR – Emissions Database for Global Atmospheric Research and others ↩︎
8. Range figures are not specified for CO₂ only in combination with LULUCF, and the values are clearly not independent – 10% is my estimate. ↩︎